Category: 3. Business

The first generation of Cayenne has achieved cult status over the years. First unveiled in 2002, the SUV represented Porsche’s bold move into a new segment and has built up a loyal fan base over the years. There are dedicated communities around the world where fans and collectors share their enthusiasm, exchange experiences and present their cars. One such modern classic recently visited Porsche’s Sonderwunsch department; Phillip Sarofim, entrepreneur and dedicated car collector, had his 2009 Cayenne GTS completely refurbished to new-car condition and extensively individualised.

This first ‘Factory Re-Commission’ project involving a Cayenne attracted a great deal of attention at Icons of Porsche. The successful combination of historic character, striking off-road looks and modern customisation was a hit. The Paint to Sample colour Black Olive created the desired retro look of the exterior. The lower sections of the body, as well as the alloy wheels, are finished in matt black. In the interior, extensive leather upholstery in English Green (Leather to Sample) meets the iconic Pasha fabric pattern in Black/Olive. The seat centres, and even the inside of the glove compartment, are trimmed in this iconic Porsche textile.

Unique Sonderwunsch Cayenne

With 70s vibes in the desert.

More customisation options than ever before with the Cayenne Electric

Right next to the reimagined classic, Porsche presented the new Cayenne Electric – for the first time in public, just days after its digital world premiere.

The technologically pioneering electric SUV, which produces up to 1,156 PS, also attracted many visitors. Its new Porsche Driver Experience display and control concept with its elegantly curved Flow Display garnered just as much attention as its impressive performance figures and progressive design did.

Another area of focus was the significantly expanded possibilities for individualisation. Never before has a Cayenne been so extensively customisable from the factory in such a wide range of ways as the new all-electric model. Customers will soon be able to choose from 13 exterior colours, 12 interior combinations and up to five interior packages and five accent packages. The new leather colours Magnesium Grey, Lavender and Sage Grey, plus leather-free Race-Tex interiors with a Pepita textile option, bring iconic Porsche designs into the present. Trim strips and accent colours can be fine-tuned to precisely match the interior and exterior.

The Cayenne goes electric

The Cayenne Electric marks the beginning of a new era for Porsche.

Together with the upgraded first-generation Cayenne in Dubai, the Cayenne Electric showcased the possibilities for individualising a Porsche using cars from more than two decades apart. The classic model demonstrates how first-generation vehicles can be transformed into bespoke one-offs through the Sonderwunsch programme. The Cayenne Electric, too, as well as offering the widest range of customisation options of any generation of Cayenne to date, can also be tailored to become a one-of-a-kind example through Porsche Exclusive Manufaktur.

Other highlights from the fifth Icons of Porsche festival

In addition to the two generations of Cayenne, Porsche also showcased other innovations and highlights at Icons of Porsche.

---

911 Turbo S

After its premiere at IAA Mobility in Munich, the new top-of-the-range model in the 911 series was also in the spotlight in Dubai. An innovative twin-turbo powertrain with T-Hybrid technology makes the new 911 Turbo S the most powerful production 911 of all time, with a system output of 523 kW (711 PS).

Macan GTS

The Macan GTS is a particularly sporty addition to Porsche’s range of all-electric SUVs. Its overboost power output of up to 420 kW (571 PS), the optional rear differential lock and the lowered sports air suspension noticeably take the agility of the model to a new level. Visually, including in the interior, the fifth variant of the Macan Electric features the characteristic GTS look with striking contrast details.

Panamera Turbo E-Hybrid Sonderwunsch

With the interior of the Panamera Turbo E-Hybrid Sonderwunsch, Porsche showcases how an individual customer dream can become a reality. Fade effects and contrasting colours that correspond to the exterior, as well as perfect craftsmanship, make the interior of this Panamera truly unique. The car’s humidor and Champagne cooler are examples of how exceptionally luxurious ideas can also be realised in a car’s interior as part of the Sonderwunsch programme.

911 GT3 with Manthey Kit

The Manthey Kit, which was previously reserved for the 911 GT3 RS, is now also available for the 911 GT3 and brings even greater performance on the racetrack. With enhanced aerodynamics offering significantly increased downforce, modified suspension, and upgraded brake components, it is designed for intensive track days. Equipped with the kit, the 911 GT3 lapped the Nürburgring-Nordschleife in 6:52.981 minutes – around 2.8 seconds faster than its predecessor fitted with the Manthey Kit.

911 Targa 4 GTS Art Car

Labubu is the most well-known character in the ‘The Monsters’ series. Together with the artist Kasing Lung, Porsche presented a strictly limited collector’s edition of the ‘King Mon’ character in Dubai as well as an art car with Labubu behind the wheel. In doing so, the collaborators celebrated 10 years of ‘The Monsters’ and 60 years of the Porsche 911 Targa at the same time. The 25th anniversary of the Carrera GT super sports car also received special attention at this year’s festival.

Introduction

Crohn’s disease (CD) is a kind of inflammatory bowel disease (IBD) characterized by transmural inflammation that can affect any segment of the gastrointestinal tract, with rising prevalence in developing countries including China.¹ Epidemiological studies reveal that 40–50% of Crohn’s disease patients develop stricture-related complications within 10 years of diagnosis, rising to 70–80% after 20 years,^2,3 while the 10-year cumulative surgical risk due to strictures reaches approximately 70%, highlighting the substantial patient and healthcare burden.⁴

While CD-related strictures can be inflammatory, fibrotic, or mixed, the accurate differentiation is critical for treatment decisions. Intestinal ultrasound (IUS) has emerged as a non-invasive tool for evaluating CD complications, with sensitivity and specificity rates of 85–95% for detecting strictures.^5,6 IUS is a noninvasive approach with several advantages, including wide availability, convenience, and low cost, and is being increasingly promoted. Recent guidelines also support its utility in diagnosing bowel strictures.⁷ However, IUS faces limitations such as operator dependency, lack of standardized protocols, and variability in equipment, despite strain elastography potentially improving accuracy in distinguishing fibrotic from inflammatory strictures.^8,9 These challenges highlight AI (artificial intelligence) ‘s potential to reduce operator bias and enhance IUS diagnostics, hence we perform this very first study to develop AI (deep learning, as automated feature extraction, and radiomics, as handcrafted features) aiding IUS-based stricture classification in CD. Radiomics captures features that are mathematically defined and can be linked to biology, while deep learning can discover complex features beyond human perception.^10,11 The previous studies have predominantly relied on CTE or MRE findings, and there is a notable lack of research on AI models based on IUS for evaluating intestinal strictures in CD. And few study conducted in endoscopic images for stricture detecting, for its risk in leading to intestinal perforation.^12,13 Moreover, existing studies are limited by small sample sizes and have primarily utilized AI to integrate clinical data with different ultrasound modalities for characterizing strictures, rather than employing true AI-based learning from image information to provide real-time diagnostic feedback.¹⁴

This single-center study aims to develop and validate radiomics and deep learning models to differentiate inflammatory and fibrotic strictures based on 87 IUS images from 64 CD patients, as the first study to apply two AI-based learning model in IUS images to differentiate fibrotic and inflammatory intestinal stricture.

Methods and Materials

Study Population

Patients with CD who underwent surgery and were hospitalized at Peking Union Medical College Hospital from January 1^st, 2018, to December 31^st, 2023, were included in the study. The specific inclusion and exclusion criteria were as follows. Inclusion criteria were: (1) CD diagnosis confirmed by Chinese IBD consensus¹⁵ and ECCO guidelines;⁵ (2) imaging evidence of strictures (luminal narrowing, bowel wall thickening, or pre-stricture dilation) via MR/CT enterography or IUS;^3,16 and (3) age ≥18 years. Exclusion criteria were: (1) non-CD strictures (eg, malignancy, ischemia); (2) recent/multiple bowel resections; (3) pregnancy/lactation. Data collected included demographics (age, sex, disease duration), clinical variables (disease location, stricture number/location, prior surgeries), medication history (biologics, corticosteroids, immunosuppressants), and surgical details (stricture location/length, time from diagnosis to surgery).

Hematoxylin & Eosin (H&E) Staining and Masson’s Trichrome Staining

Resected bowel specimens from CD patients are fixed in 10% neutral buffered formalin for 24 hours, embedded in paraffin, and sectioned into 4–5 μm slices. The most stenotic intestinal segment and adjacent areas are selected for staining. For H&E staining, sections are deparaffinized, rehydrated, stained with hematoxylin and eosin, dehydrated, cleared, and mounted. For Masson’s trichrome staining, sections undergo similar preparation, followed by staining with Weigert’s iron hematoxylin (nuclei), Biebrich scarlet-acid fuchsin (muscle/cytoplasm), and aniline blue (collagen). Slides are differentiated, dehydrated, cleared, and cover-slipped. 3D histech captured representative images, and ImageJ (Version 1.53. US National Institutes of Health, https://imagej.net) quantified collagen-stained tissue relative to total tissue area. The fibrosis staining area ratio was calculated for each specimen, and the median ratio was determined. Patients were divided into two groups based on the median: those with Masson staining area above (severe fibrosis) or below the median. This binary classification aligns with methods used in prior literature.^17,18 The diagnosing pathologist had access to clinical and imaging data necessary for standard diagnostic workflow but was blinded to the experimental ultrasound classifications. A second pathologist, performing the quantitative Masson’s trichrome staining analysis, was fully blinded to all imaging data and calculated the collagen area ratio based solely on the HE-stained slides to ensure objective, research-specific assessment.

Ultrasonographic Examination and Definition of the ROI

Intestinal ultrasound (IUS) was conducted following the European Federation of Societies for Ultrasound in Medicine and Biology guidelines and Experts suggestions on the standardization of intestinal ultrasound examination and reporting for inflammatory bowel disease in China.¹⁹ A standardized, comprehensive intestinal scan was performed by one of three radiologists, each with over 10 years of experience, using a Philips iU22 (Philips Healthcare, Bothell, WA, USA) or SuperSonic Aixplorer (SuperSonic Imaging, SA, France) machine equipped with convex (C5-2) and linear (L9-3) transducers. Patients fasted for at least 8 h before US examination followed the instruction of gastroenterologists. And most of them typically follow a low-residue diet within a few days before screening or due to stricture situation, hence minimizing the interference caused by intestinal contents. A thorough scanning of the colon (from the ileocecal region to the sigmoid) and small intestine was performed with the convex transducer first. Then a detailed examination was performed by the linear transducer.

So far there is no universally accepted threshold for diagnosing intestinal stricture. The Stenosis Therapy and Anti-Fibrotic Research Consortium recommends the following ultrasound criteria for diagnosing small intestinal stricture in CD: bowel wall thickness ≥3–4 mm, narrowed luminal diameter (<1 cm), accompanied by proximal bowel loop dilation (>2.5 cm).²⁰ Once diagnosed with intestinal stricture, the stricture segments are classified into three categories based on following ultrasonic features: fibrotic stricture, inflammatory stricture, and mixed-type. Fibrotic strictures are defined as distinct bowel wall stratification with minimal or absent vascularity, regardless of the bowel wall thickness. Inflammatory strictures are defined as loss of bowel wall stratification with long stretches of vascularity or vascularity reaching the mesentery regardless of the bowel wall thickness; or indistinct bowel wall stratification, long stretches of vascularity reaching the mesentery, regardless of the bowel wall thickness. For the clinical strategy and the characterization of US qualifications, the diagnostic value was calculated by comparing the fibrotic stricture group with the inflammatory stricture group.

Two radiologists (QJ, >5 years’ experience; ZQL, >10 years’ experience) independently reviewed the images, blinded to clinical, laboratory, and histopathological data. Disagreements were resolved by a third radiologist (WB.L, >10 years’ experience). Regions of interest (ROIs) were manually delineated on representative stricture images by a radiologist (MY.Z, 5 years’ experience), blinded to histopathological findings, using ImageJ software.

Radiomics-Based Classification Method

The most representative images from each patient in the original dataset were selected, resulting in 87 images cropped to target regions. We used 5-fold cross-validation, where in each fold, the training set was used for feature extraction and diagnostic model construction, and the test set was used to evaluate model performance. The final results were averaged across the five folds to represent the overall model performance. Radiomics features are extracted using the PyRadiomics library in Python, enabling quantification of various imaging characteristics. All images were normalized to zero mean and unit variance prior to feature extraction to reduce inter-subject intensity variation. Image preprocessing followed the PyRadiomics configuration “binWidth” = 25, “force2D” = True, “interpolator” = sitk.sitkBSpline, “resampledPixelSpacing” = None. Here, a fixed gray-level bin width of 25 was used for intensity discretization. The force2D = True option ensured 2D feature extraction, consistent with the image dimensionality. B-spline interpolation was applied for any necessary image resampling to achieve smooth voxel transitions, while resampledPixelSpacing = None retained the native resolution when voxel spacing was uniform. ROIs were generated from doctor’s annotation and verified using imageoperations.checkMask to ensure spatial alignment. The extracted features include first-order statistical features (eg, mean, standard deviation, etc)., shape features (eg, volume, area, etc.), texture features [such as grayscale covariance matrices (GLCM), grayscale tour length matrices (GLRLM), and grayscale size-zone matrices (GLSZM) etc.]. Over 1100 radiomics features were calculated, with feature selection based on Pearson correlation and statistical significance (p < 0.05). A random forest classifier was employed to differentiate inflammatory from fibrotic strictures (Figure 1A). Model performance was assessed using accuracy, sensitivity, specificity, positive/negative predictive values, F1 score, confusion matrix, and AUC.

Figure 1 (A) The framework of the radiomics-based approach; (B) The framework for classifying inflammatory or fibrotic strictures based on the Resnet50 model.

Deep Learning-Based Classification Model

ResNet50, a deep learning model for image recognition and classification, is a representative residual network designed to address gradient vanishing and network degradation in deep architectures. It enhances training efficiency through residual blocks, which facilitate rapid information transfer across layers. The model comprises convolutional layers, residual blocks, pooling layers, and fully connected layers. Convolutional layers extract features from input images, while residual blocks use shortcut connections to preserve and propagate information. Pooling layers reduce feature map dimensionality, mitigating overfitting and computational complexity. The fully connected layer outputs class scores, converted into probability distributions via the Softmax activation function (Figure 1B).

The ResNet50 model was subjected to 5-fold cross-validation using the same 87 images dataset used for radiomics analysis. To ensure strict test set independence, all images from the same patient were assigned to the same fold. Before training, all ultrasound images were first cropped to include only the ROI to focus on diagnostically relevant structures and minimize background noise. Subsequently, the images underwent a standardized preprocessing procedure. Each image was resized to 128*128 pixels to ensure consistent input dimensions across the dataset, and normalized to have a mean and standard deviation of 0.5 for each channel. Training employed an initial learning rate of 0.0001, adjusted periodically using step-down scheduling, with a batch size of 16 and 100 epochs. Model performance was evaluated on the test set using metrics including accuracy, sensitivity, specificity, positive/negative predictive values, F1 score, confusion matrix, and AUC.

Implementation Details Subsection

All experiments were conducted on an Ubuntu 24.04 operating system using a single Nvidia GeForce RTX 3090 GPU. The programming environment was based on Python version 3.8.19, and the deep learning framework used was PyTorch version 2.0.1.

Ethic Statement

Based on the Ethics Committee Guidelines of Peking Union Medical College Hospital for the Clinical Research Involving Human Subjects, this study is exempt from obtaining informed consent from the subjects, as such exemption does not negatively impact their rights and interests. The research utilizes identifiable human materials or data where the subjects can no longer be located, and the study does not involve personal privacy or commercial interests. Following review by the Ethics Committee (I-22PJ1092), this retrospective study has been deemed to meet the above criteria and is therefore exempt from the requirement of signed informed consent. And the study complies with the Declaration of Helsinki.

Sample Size and Statistical Analysis

The sample size was calculated based on an expected sensitivity of 80% for the diagnostic model,¹⁴ with a 95% confidence level and a confidence interval width of 0.20, yielding a minimum requirement of 61 patients. Based on this estimation for model sensitivity, we retrospectively enrolled 64 surgically confirmed CD patients from our institutional database who had undergone preoperative intestinal ultrasound. Continuous variables with a normal distribution are expressed as the mean ± standard deviation (SD), while nonnormal variables are reported as medians (interquartile ranges [IQRs]). Categorical and discrete variables are presented as percentages. The means of two continuous normally distributed variables were compared using Student’s t-test, and the Mann–Whitney U-test was applied to compare nonnormally distributed variables. The frequencies of categorical variables were compared using Pearson or Fisher’s exact test under specific conditions. P < 0.05 was considered to indicate statistical significance. Cohen’s kappa (κ) coefficient was used to assess the agreement between the IUS findings of the two observers (QL.Z. and J.Q). The level of agreement was defined as poor (κ < 0.20), fair (0.2 < κ ≤ 0.40), moderate (0.4 < κ ≤ 0.60), good (0.6 < κ ≤ 0.80) and very good (0.8 < κ ≤ 1.0). All the statistical analyses were performed using SPSS (version 23.0; SPSS Inc., Chicago, IL, USA).

Results

Baseline Data for the Included Patients with CD

This study included 64 CD patients, with a median Masson’s staining area of 40.10% (IQR: 35.55%-41.96%). The baseline characteristics of CD patients were presented in Supplementary Table S1, grouped by Masson staining ratio, with significant differences observed between the groups (33.25% vs 47.29%, P = 0.037). Of 34 patients with small intestinal strictures, 24 (70.59%) were fibrotic based on Masson’s staining; among 30 patients with colonic strictures, 17 (56.66%) were fibrotic. The high Masson staining group had a longer disease duration (9.65 ± 2.43 vs 7.66 ± 2.12 years, P = 0.040) and a longer interval from diagnosis to surgery (8.90 ± 3.12 vs 6.22 ± 1.41 years, P = 0.047) compared to the low staining group.

Model Performance on an Internal Independent Test Set

Radiomics-Based Method

In our experiments, we first classified intestinal stricture as fibrosis and inflammatory based on radiomics-based method. During training, using radiomics-based method, the classification has the accuracy of 67.0% (95% Confidence Interval [CI], 44.4%–88.9%), a sensitivity of 75.0% (95% CI, 40.0–100%), a specificity of 60.0% (95% CI, 28.6%–90.0%), a positive predictive value of 60.0% (95% CI, 27.3%–90.0%), a 75.0% (95% CI, 42.9%–100%) for negative predictive value, 67.0% (95% CI, 33.3%–90.0%) for F1 score, and 67.5% for AUC. We also show the importance of the features extracted by the radiomics-based approach. Figure 2 shows the importance of features. Among that, the top 10 important features are logarithm_InverseVariance, ShortRunLowGrayLevelEmphasis, gradient_SmallAreaHighGrayLevelEmphasis, wavelet-HLL_RunLengthNonUniformity, gradient_Autocorrelation, gradient_ShortRunLowGrayLevelEmphasis, gradient_RunLengthNonUniformity, wavelet-HLH_SmallAreaEmphasis, gradient_Idm, and wavelet-HLH_Energy. The radiomics methods help improve the interpretability of our model.

Figure 2 Importance of radiomics features.

Deep Learning-Based Method

Training with the Resnet50 model yielded the classification accuracy of 83.8% (95% CI, 66.7%–100%), sensitivity of 88.9% (95% CI, 62.5%–100%), specificity of 77.8% (95% CI, 49.9%–100%), positive predictive value of 80.0% (95% CI, 54.6%–100%), negative predictive value 87.5% (95% CI, 60.0%–100%), F1 score of 84.2% (95% CI, 61.5%–100%), and AUC of 70.0%.

Deep learning performs better compared to radiomics in our experiments. This is because radiomics relies on manual feature extraction, while deep learning uses automatic feature extraction. This can automatically extract complex and highly abstract features from data through the multi-layer neural networks, avoiding the limitations of manual intervention and showing stronger generalization ability and robustness.

We also show the confusion matrix results of both radiomics and deep learning-based methods in Figure 3. In the confusion matrices, the rows represent the true categories of the test images, the columns represent the predicted categories of the test images (Negative for inflammatory stricture and Positive for fibrotic stricture). Obviously, the confusion matrix of the Resnet50 model (Figure 3B) shows a clear advantage on the diagonal compared with the radiomics-based method (Figure 3A). As for the ROC results, we can observe that the ROC curve for the Resnet50 model (Figure 3C) shows a trend of gradually approaching the upper left corner. This shows that the model exhibits better classification ability than the radiomics-based method (Figure 3C). The deep learning model demonstrated a significantly higher AUC compared to the radiomics model (P = 0.018). The deep learning model also showed a statistically significant superiority over the expert assessments (P = 0.043). The difference between the radiomics model and the expert assessments was not statistically significant (P = 0.271) (Figure 3C).

Figure 3 Testing the performance of (A) Radiomics and (B) Resnet50 models using confusion matrices (The vertical axis represents the true labels and the horizontal axis represents the model’s predicted labels, with fibrotic indicated by positive and inflammatory indicated by negative); (C) testing the Performance of Resnet50 Models, Radiomics and experts’assessment based on ROC Curves.

Visualization

The model’s prediction results are visualized in Figure 4. Panel A displays fibrotic stricture predictions: the first three images are correctly classified, while the fourth misclassifies a fibrotic stricture as inflammatory. Panel B shows inflammatory stricture predictions: the first three are accurate, and the fourth incorrectly labels an inflammatory stricture as fibrotic.

Figure 4 Visualization of attention scores for successful cases in combination with deep learning models. Representative images of patients predicted as (A) Fibrotic and (B) Inflammatory are visualized to illustrate the prediction process of the Resnet50 model.

Attentional Visualization of Deep Learning Models

To enhance interpretability, we employed class activation maps (CAMs) for attention visualization in deep learning models. CAMs, matching the original image size, assign pixel values from 0 to 1 (grayscale: 0–255), representing the contribution to the predicted output. Higher scores indicate greater sensitivity and network contribution from corresponding image regions. In our study, CAMs were generated for images in Figure 5, with heatmaps visualizing neural network features. The first row displays original images, while the second row overlays CAMs. Red areas highlight key discriminative regions, with intensity reflecting feature effectiveness.

Figure 5 Resnet50 model class activation maps: (A) real label fibrotic, and predicted label as fibrotic; (B) real label as inflammatory, predicted label as fibrotic; (C) real label as fibrotic, predicted label as inflammatory; (D) real label as inflammatory, predicted label as inflammation.

Results demonstrate the model’s ability to precisely focus on critical areas during intestinal stricture classification, effectively identifying abnormal features in the images. This underscores the model’s capability in targeting relevant pathological regions for accurate classification.

Comparative Analysis of Radiomics, Resnet50 Model, and Expert Predictions

The ResNet50 model achieved the highest accuracy (83.3%), surpassing Radiomics (67.0%) and expert radiologists (73.1%) (Table 1). It also led in sensitivity (88.9% vs 75.0% for Radiomics and 54.6% for experts) and positive predictive value (PPV: 80.0% vs 75.0% for experts and 60.0% for Radiomics). Experts demonstrated the highest specificity (86.7% vs 77.8% for ResNet50 and 60.0% for Radiomics). The ResNet50 model also excelled in negative predictive value (NPV: 87.5%) and F1-score (84.2%), outperforming experts (NPV: 72.2%; F1-score: 69.7%) and Radiomics (NPV: 60.0%; F1-score: 67.0%). While ResNet50 consistently outperformed in most metrics, expert specificity remained superior. However, expert performance in stricture classification was suboptimal overall. Inter-observer agreement between the two experts was very good (Cohen’s κ = 0.815; p < 0.001). These findings highlight ResNet50’s potential as a robust tool for medical image analysis.

Table 1 Comparative Analysis of Diagnostic Performance Metrics Among Radiomics, ResNet50 Model, and Experienced Expert in Classifying Intestinal Stricture

Discussion

This study presents several significant findings regarding the clinical characteristic in CD patients and classification of intestinal strictures among them using AI approaches. Our deep learning-based method (Resnet50) achieved superior performance compared to the radiomics-based approach and expert predictions, with higher accuracy and better sensitivity in distinguishing inflammatory from fibrotic strictures. However, expert predictions achieved better specificity among them. Meanwhile, the CAMs visualization demonstrated that the deep learning model could effectively identify and focus on relevant pathological features, enhancing the model’s interpretability and clinical applicability. AI-driven analysis into routine clinical practice holds the promise of standardizing interpretations of IUS results, thus reducing inter-observer variability.

While several AI models have been developed for differentiating Crohn’s disease–related strictures, many prior studies have primarily utilized CTE or MRE. For instance, one model based on radiologist-defined strictures using automated CTE measurements achieved an accuracy of 87.6%.²¹ Another deep learning approach outperformed two radiologists (AUCs: 0.579 and 0.646; both P < 0.05) and was not inferior to a radiomics model (AUC = 0.813, P < 0.05), while requiring significantly less processing time (P < 0.001).²² Our model achieved comparable accuracy. Considering the operator-dependent nature of intestinal ultrasound and the relatively smaller sample size of our study compared to previous CTE/MRE-based studies, these results suggest that our model performs similarly to existing approaches.

The superior performance of deep learning over traditional radiomics highlights its potential in medical imaging analysis.^23–25 This study is the first to compare radiomics and deep learning in differentiating CD strictures using ultrasound. Notably, expert predictions demonstrated the highest specificity, underscoring their continued importance despite AI advancements.

The successful implementation of attention visualization through CAMs represents a significant step toward interpretable AI in clinical practice. This addresses a crucial concern in medical AI applications – the “black box” nature of deep learning models. Recent work has similarly emphasized the importance of interpretable AI in clinical decision-making, showing that visualization techniques can increase physician trust and adoption of AI systems.²⁶ This automatic focusing ability not only improves the interpretability of the model but also makes the deep learning model more consistent in comparison with expert evaluation, enhancing the credibility and interpretability of the model in practical applications. To enhance generalizability, a standardized intestinal ultrasound imaging protocol should first be established to ensure consistent and effective image acquisition for AI-assisted diagnosis. In this study, the radiologists strictly followed the screening protocol recommended by Experts suggestions on the standardization of intestinal ultrasound examination and reporting for inflammatory bowel disease in China.¹⁹ And two types of ultrasound machines were applied during the study (Philips iU22 or SuperSonic Aixplorer). Both of them showed similar performance in recognizing the characteristic of fibrotic or inflammatory stricture. Also, CD patients strictly fast for 8 hours prior to the examination followed the instruction of gastroenterologists, and most of them typically follow a low-residue diet, hence minimizing the interference caused by intestinal contents. And the radiologists adopted the unified diagnosis criteria of intestinal stricture for patients inclusion.²⁰ Based on a unified patient preparation protocol, quantitative criteria for intestinal stenosis assessment, and a standardized screening and image acquisition process, the model established herein demonstrates strong potential for generalizability following the broad implementation of these standardized examination procedures. Future validation should explicitly evaluate model performance across diverse equipment from various manufacturers and models to assess true portability. Furthermore, a prospective multi-center study specifically designed to test the model across different healthcare settings, ultrasound machines, and operator skill levels would significantly strengthen its general applicability.

The integration of ultrasound-based AI analysis into CD management could significantly impact clinical practice. Recent studies emphasize the growing role of IUS in CD monitoring, demonstrating high sensitivity, specificity, and concordance with endoscopic scores.²⁷ IUS is recommended as an ideal diagnostic tool for long-term follow-up of IBD according to both China and ECCO guidelines^28,29 Our AI approach could enhance the utility of this non-invasive imaging modality by providing objective, quantitative assessment of stricture characteristics. Using deep neural networks can perfectly tell difference between strictures and normal mucosa (AUC = 0.989), as well as strictures and all ulcers (AUC = 0.942) based on capsule endoscopy images.³⁰ For patients intolerant to endoscopy, IUS offers a safer alternative for stricture evaluation.

The study demonstrates several notable strengths in both methodology and execution. First, it employs a sophisticated dual-approach methodology combining both radiomics and deep learning techniques, which provides a comprehensive framework for analyzing intestinal strictures in CD patients. Second, the inclusion of CAM for visualization adds a crucial layer of interpretability to the deep learning model, making the results more transparent and clinically applicable.

Despite its strengths, the study has several limitations that warrant consideration. Although this study represents the largest sample size to date for assessing intestinal stricture characteristics using intestinal ultrasound, its single-center design limits the generalizability of the findings. Additionally, the split ratio of 8:2 for training and testing sets, while common in machine learning studies, might not provide sufficient test data for robust validation given the small overall sample size. Another limitation is the lack of external validation on an independent dataset from different medical centers, which would be crucial for establishing the model’s generalizability. The study also does not address the potential impact of different ultrasound equipment and operators on image quality and subsequent analysis, which could be a significant source of variability in real-world applications. In the future, as a leading center for IBD diagnosis in China, we aim to standardize and promote intestinal ultrasound practices across multiple hospitals. Future efforts will focus on expanding the deep learning model to multicenter settings, incorporating larger-scale datasets, and closely tracking patient outcomes. This will enhance the applicability of our findings and allow more patients to benefit from the research. To enhance clinical integration and model interpretability, we propose the following steps: First, the AI could be trained to recognize guideline-recommended sonographic features indicative of fibrotic or inflammatory strictures—features identifiable by sonographers but subject to interpreter experience. Second, the model’s accuracy, sensitivity, and specificity should be validated through larger, multi-center studies. Third, technical integration with ultrasound systems should be pursued to enable real-time feedback during examinations. Finally, standardized imaging protocols must be widely promoted to support consistent application.

Conclusion

This pioneering study compares radiomics and deep learning for differentiating fibrotic stricture from inflammatory stricture in CD patients, highlighting the superior performance of the ResNet50 model in accuracy and diagnostic metrics. Regarding the rather small sample size and lack of multi-center data, future multi-center studies with external validation and longitudinal data are needed to assess generalizability and predictive capabilities for disease progression. Prospective studies incorporating the clinical data and IUS images, with essential follow-up, will be performed to validate clinical utility in real-world settings.

Data Sharing Statement

All data and material are shown in this manuscript.

Ethics Approval and Informed Consent

Based on the Ethics Committee Guidelines of Peking Union Medical College Hospital for the Clinical Research Involving Human Subjects, this study is exempt from obtaining informed consent from the subjects, as such exemption does not negatively impact their rights and interests. The research utilizes identifiable human materials or data where the subjects can no longer be located, and the study does not involve personal privacy or commercial interests. Following review by the Ethics Committee (I-22PJ1092), this retrospective study has been deemed to meet the above criteria and is therefore exempt from the requirement of signed informed consent. And the study complies with the Declaration of Helsinki.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This work was supported by National Key R&D Program of China (2023YFC2507300), Beijing Health Technology Promotion Project (BHTP P2024096, BHTPP P2024097), National High-Level Hospital Clinical Research Funding (2025-PUMCH-A-163, 2022-PUMCH-B-022, 2022-PUMCH-C-018), CAMS Innovation Fund for Medical Sciences (2024-I2M-C&T-B-004), and State Key Laboratory Special Fund (2060204).

Disclosure

The authors report no conflicts of interest in this work.

References

1. Little RD, Jayawardana T, Koentgen S, et al. Pathogenesis and precision medicine for predicting response in inflammatory bowel disease: advances and future directions. eGastroenterology. 2024;2(1):e100006. doi:10.1136/egastro-2023-100006

2. Gordon IO, Agrawal N, Willis E, et al. Fibrosis in ulcerative colitis is directly linked to severity and chronicity of mucosal inflammation. Aliment Pharmacol Ther. 2018;47(7):922–939. doi:10.1111/apt.14526

3. Rieder F, Ma C, Hanzel J, et al. Reliability of CT enterography for describing fibrostenosing Crohn disease. Radiology. 2024;312(2):e233038. doi:10.1148/radiol.233038

4. Thia KT, Sandborn WJ, Harmsen WS, Zinsmeister AR, Loftus EV. Risk factors associated with progression to intestinal complications of Crohn’s disease in a population-based cohort. Gastroenterology. 2010;139(4):1147–1155. doi:10.1053/j.gastro.2010.06.070

5. Gomollón F, Dignass A, Annese V, et al. 3rd European evidence-based consensus on the diagnosis and management of Crohn’s disease 2016: part 1: diagnosis and medical management. J Crohn’s Colitis. 2017;11(1):3–25. doi:10.1093/ecco-jcc/jjw168

6. Fraquelli M, Castiglione F, Calabrese E, Maconi G. Impact of intestinal ultrasound on the management of patients with inflammatory bowel disease: how to apply scientific evidence to clinical practice. Digestive Liver Dis. 2020;52(1):9–18. doi:10.1016/j.dld.2019.10.004

7. Lu C, Rosentreter R, Parker CE, et al. International expert guidance for defining and monitoring small bowel strictures in Crohn’s disease on intestinal ultrasound: a consensus statement. Lancet Gastroenterol Hepatol. 2024;9(12):1101–1110. doi:10.1016/S2468-1253(24)00265-6

8. Chen YJ, Mao R, Li XH, et al. Real-time shear wave ultrasound elastography differentiates fibrotic from inflammatory strictures in patients with Crohn’s disease. Inflammatory Bowel Dis. 2018;24(10):2183–2190. doi:10.1093/ibd/izy115

9. Pescatori LC, Mauri G, Savarino E, Pastorelli L, Vecchi M, Sconfienza LM. Bowel sonoelastography in patients with Crohn’s disease: a systematic review. Ultrasound Med Biol. 2018;44(2):297–302. doi:10.1016/j.ultrasmedbio.2017.10.004

10. Turco S, Tiyarattanachai T, Ebrahimkheil K, et al. Interpretable machine learning for characterization of focal liver lesions by contrast-enhanced ultrasound. IEEE Trans Ultrason Ferroelectr Freq Control. 2022;69(5):1670–1681. doi:10.1109/TUFFC.2022.3161719

11. Whitney HM, Li H, Ji Y, Liu P, Giger ML. Comparison of breast MRI tumor classification using human-engineered radiomics, transfer learning from deep convolutional neural networks, and fusion methods. Proceed IEEE Institute Electrical Electronics Engineers. 2020;108(1):163–177. doi:10.1109/JPROC.2019.2950187

12. Maeda Y, Ditonno I, Puga-Tejada M, et al. Artificial intelligence-enabled advanced endoscopic imaging to assess deep healing in inflammatory bowel disease. eGastroenterology. 2024;2(3):e100090. doi:10.1136/egastro-2024-100090

13. Majtner T, Brodersen JB, Herp J, Kjeldsen J, Halling ML, Jensen MD. A deep learning framework for autonomous detection and classification of Crohn’s disease lesions in the small bowel and colon with capsule endoscopy. Endoscopy Int Open. 2021;9(9):E1361–e1370. doi:10.1055/a-1507-4980

14. Chen YF, Liu L, Lyu B, et al. Role of artificial intelligence in Crohn’s disease intestinal strictures and fibrosis. J Digestive Dis. 2024;25(8):476–483. doi:10.1111/1751-2980.13308

15. Inflammatory Bowel Disease Group CSoGCMA, China IBDQCCo. Chinese clinical practice guideline on the management of Crohn′s disease (2023, Guangzhou). Chin J Inflamm Bowel Dis. 2024;08(1):2–32.

16. Bettenworth D, Baker ME, Fletcher JG, et al. A global consensus on the definitions, diagnosis and management of fibrostenosing small bowel Crohn’s disease in clinical practice. Nat Rev Gastroenterol Hepatol. 2024;21(8):572–584. doi:10.1038/s41575-024-00935-y

17. Zhang MC, Li XH, Huang SY, et al. IVIM with fractional perfusion as a novel biomarker for detecting and grading intestinal fibrosis in Crohn’s disease. Eur Radiol. 2019;29(6):3069–3078. doi:10.1007/s00330-018-5848-6

18. Mao H, Su P, Qiu W, Huang L, Yu H, Wang Y. The use of Masson’s trichrome staining, second harmonic imaging and two-photon excited fluorescence of collagen in distinguishing intestinal tuberculosis from Crohn’s disease. Colorectal Dis. 2016;18(12):1172–1178. doi:10.1111/codi.13400

19. Chinese Quality Control Assessment Center for Inflammatory Bowel Disease Diagnosis and Treatment IBDG, Chinese Society of Gastroenterology, Chinese Medical Association;, Abdominal Ultrasound Group, Chinese Society of Ultrasound in Medcine, Chinese Medical Association. Experts suggestions on the standardization of intestinal ultrasound examination and reporting for inflammatory bowel disease in China. Chin J Inflamm Bowel Dis. 2024;08(2):109–115.

20. Bettenworth D, Bokemeyer A, Baker M, et al. Assessment of Crohn’s disease-associated small bowel strictures and fibrosis on cross-sectional imaging: a systematic review. Gut. 2019;68(6):1115–1126. doi:10.1136/gutjnl-2018-318081

21. Stidham RW, Enchakalody B, Waljee AK, et al. Assessing small bowel stricturing and morphology in Crohn’s disease using semi-automated image analysis. Inflammatory Bowel Dis. 2020;26(5):734–742. doi:10.1093/ibd/izz196

22. Meng J, Luo Z, Chen Z, et al. Intestinal fibrosis classification in patients with Crohn’s disease using CT enterography-based deep learning: comparisons with radiomics and radiologists. Eur Radiol. 2022;32(12):8692–8705. doi:10.1007/s00330-022-08842-z

23. Song D, Zhang Z, Li W, Yuan L, Zhang W. Judgment of benign and early malignant colorectal tumors from ultrasound images with deep multi-View fusion. Comput Methods Programs Biomed. 2022;215:106634. doi:10.1016/j.cmpb.2022.106634

24. Bedrikovetski S, Dudi-Venkata NN, Kroon HM, et al. Artificial intelligence for pre-operative lymph node staging in colorectal cancer: a systematic review and meta-analysis. BMC Cancer. 2021;21(1):1058. doi:10.1186/s12885-021-08773-w

25. Bao Z, Du J, Zheng Y, Guo Q, Ji R. Deep learning or radiomics based on CT for predicting the response of gastric cancer to neoadjuvant chemotherapy: a meta-analysis and systematic review. Front Oncol. 2024;14:1363812. doi:10.3389/fonc.2024.1363812

26. van der Velden BHM, Kuijf HJ, Gilhuijs KGA, Viergever MA. Explainable artificial intelligence (XAI) in deep learning-based medical image analysis. Med Image Anal. 2022;79:102470. doi:10.1016/j.media.2022.102470

27. Madsen GR, Wilkens R, Boysen T, et al. The knowledge and skills needed to perform intestinal ultrasound for inflammatory bowel diseases-an international Delphi consensus survey. Aliment Pharmacol Ther. 2022;56(2):263–270. doi:10.1111/apt.16950

28. Treatment CQCACfIBDDa. Experts suggestions on the standardization of intestinal ultrasound examination and reporting for inflammatory bowel disease in China. Chin J Inflamm Bowel Dis. 2024;08(8):109–115.

29. Kucharzik T, Tielbeek J, Carter D, et al. ECCO-ESGAR topical review on optimizing reporting for cross-sectional imaging in inflammatory bowel disease. J Crohn’s Colitis. 2022;16(4):523–543. doi:10.1093/ecco-jcc/jjab180

30. Klang E, Grinman A, Soffer S, et al. Automated detection of Crohn’s disease intestinal strictures on capsule endoscopy images using deep neural networks. J Crohn’s Colitis. 2021;15(5):749–756. doi:10.1093/ecco-jcc/jjaa234

This article picked by a teacher with suggested questions is part of the Financial Times free schools access programme. Details/registration here.

Specification:

Some current statistics about US monetary policy and unemployment:

The current rate of unemployment in the US is 4.4 per cent, the third lowest in the G7 behind Japan (2.56 per cent) and Germany (3.41 per cent)

The US inflation rate of 3 per cent is close to its target rate, but the second-highest in the G7.

Unemployment is measured by the US Bureau of Labor Statistics

US President Donald Trump has long campaigned for the Fed to cut rates, arguing that high interest rates act as a barrier to jobs and investment. But the Federal Reserve is independent and must consider a range of factors when deciding to cut rates,

Read the article and then answer the questions:

US adds 119,000 jobs in September but unemployment hits four-year peak

• Define the term unemployment [2]

• Outline the current trend in US unemployment [2]

• Using an aggregate demand and supply diagram, explain the impact of the US central (Federal Reserve) decreasing its base interest rate on US unemployment [4]

• Explain two other macroeconomic factors that the Federal Reserve might consider when deciding whether to decrease its base interest rate [4]

• Discuss the effectiveness of decreasing interest rates to reduce unemployment [15]

Mark Johnson, InThinking/thinkIB

Bioinformatic analysis of PRMT1 expression and clinical correlation using TCGA data

Publicly available RNA sequencing (RNA-seq) data (Level 3 HTSeq—Counts and Fragments Per Kilobase of transcript per Million mapped reads [FPKM]) and associated clinical information, including overall survival data, for the Head and Neck Squamous Cell Carcinoma (HNSC) cohort were accessed and downloaded from The Cancer Genome Atlas (TCGA) database portal (https://portal.gdc.cancer.gov/). Data processing and normalization pipelines adhered to TCGA standards. For differential expression analyses comparing tumor versus adjacent normal tissues, normalized expression values (e.g., Transcripts Per Million [TPM]) were used. Where appropriate for statistical testing, Log2 transformation (log2[TPM + 1] or log2[FPKM + 1]) was applied to the expression data to approximate a normal distribution.

The OSCC subset within of TCGA-HNSC cohort was identified based on primary tumor site annotations (e.g., oral cavity, tongue, floor of mouth) within the clinical data, and only these designated samples were included in the subsequent OSCC-specific analyses. To compare PRMT1 mRNA expression between tumor and adjacent normal tissues, we utilized publicly accessible online tools integrating TCGA data were employed, primarily the Gene Expression Profiling Interactive Analysis (GEPIA) web server (http://gepia.cancer-pku.cn/) and/or the UALCAN portal (http://ualcan.path.uab.edu/).

Survival analysis

To assess the prognostic significance of PRMT1 expression in OSCC, survival analysis was conducted using the clinical follow-up data linked to the RNA-seq profiles of the TCGA-OSCC patient subset. This analysis utilized integrated tools within GEPIA or cBioPortal (http://www.cbioportal.org/), or was performed using custom scripts in R (version 3.6) with the survival and survminer packages. OSCC patients were stratified into “High PRMT1 expression” and “Low PRMT1 expression” groups. The stratification cutoff was determined by the median expression value across the OSCC cohort, though optimal cutoffs determined by the platform was considered. The specific cutoff method used (e.g., median, quartile) was noted from the analysis output. Kaplan–Meier survival curves were generated to visualize the overall survival probability over time for the high- and low-PRMT1 expression groups. The statistical significance of the difference in overall survival between the two groups was assessed using the log-rank (Mantel-Cox) test. A P-value < 0.05 was considered indicative of a statistically significant difference in survival outcomes associated with PRMT1 expression levels. Hazard ratios (HR) and 95% confidence intervals were also recorded from the analysis tools.

Ethical approval and tissue sample collection

This study received approval from the Ethics Committee of Chinese PLA General Hospital (S2025-018–01). The study was conducted according to the guidelines of the Declaration of Helsinki. Written informed consent was obtained from all participating patients prior to sample collection. OSCC tissues, categorized by grade (Grade I, Grade II, Grade III) and lymph node metastasis status (with LN metastasis, without LN metastasis), along with adjacent non-cancerous tissues, were surgically procured from patients undergoing treatment at Chinese PLA General Hospital. Immediately following resection, tissue samples were snap-frozen in liquid nitrogen and stored for subsequent molecular analyses.

Immunohistochemistry (IHC)

Four-micrometer thick sections were prepared from paraffin-embedded blocks human OSCC tumor tissues. The sections were deparaffinized in xylene and rehydrated through a graded ethanol solutions. Heat-mediated antigen retrieval was performed in an appropriate buffer solution. Following blocking of non-specific binding sites, sections were incubated overnight at 4 °C with a primary antibody against PRMT1 (1:500 dilution; ab190892; Abcam, Shanghai, China). After washing, sections were incubated with an appropriate horseradish peroxidase (HRP)-conjugated secondary antibody (ab6721, 1:1000 dilution; Abcam, Shanghai, China) for 2 h at room temperature. Visualization was achieved using a 3,3′-Diaminobenzidine (DAB) substrate kit, followed by counterstaining with hematoxylin. Stained sections were dehydrated, cleared, and mounted. Images were captured using a light microscope (Nikon, Tokyo, Japan). For immunohistochemical analysis of Vimentin and E-cadherin in xenograft tumors, similar procedures were followed using respective primary antibodies after tissue processing.

Cell lines and culture conditions

Human OSCC cell lines HN6, SCC25, Cal27, and SCC15, along with normal human oral keratinocytes (HOK), were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). All cell lines were maintained in Dulbecco’s Modified Eagle Medium (DMEM; Gibco, USA) supplemented with 10% fetal bovine serum (FBS; Gibco Laboratories, USA) and 1% penicillin–streptomycin. Cells were cultured in a humidified incubator at 37 °C with an atmosphere containing 5% CO₂.

Plasmid construction and cell transfection

Expression vectors based on pcDNA3.1 for overexpressing PRMT1 (OE-PRMT1) and a corresponding empty vector control (OE-NC) were also obtained from GenePharma. For transient transfections, HN6 and SCC25 cells were seeded to reach appropriate confluency. Plasmids were transfected into the cells using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. Cells were harvested 48 h post-transfection for subsequent experiments.

Lentiviral shRNA knockdown and generation of stable cell lines

shRNA oligonucleotides targeting human PRMT1 and STAT3 (two independent hairpins per gene) and a non-targeting control (sh-NC) were cloned into pLKO.1-puro. Lentiviruses were produced by co-transfecting HEK293T cells with the shRNA vector and packaging plasmids (pLP1, pLP2, pLP/VSVG; Invitrogen) using Lipofectamine 2000 according to the manufacturer’s instructions. Viral supernatants were collected at 48 and 72 h, clarified by 0.45-µm filtration, supplemented with 4 µg/mL polybrene, and used to infect HN6 or SCC15 cells (two 24-h rounds, MOI ~ 2–5). Forty-eight hours after the final infection, cells were selected in puromycin (2 µg/mL) for 2–3 days to generate stable pooled populations. Knockdown efficiency was assessed by qRT-PCR (Supplementary Fig. S1a–b) and immunoblotting (Fig. 2a). For subsequent experiments we used sh-PRMT1-1 and sh-STAT3-2, which showed the strongest silencing, while results for both hairpins are reported to exclude off-target effects.

Western blot analysis

Total protein was extracted from cultured OSCC cells using RIPA lysis buffer supplemented with protease and phosphatase inhibitors. Protein concentrations were determined using the BCA protein assay kit (Beyotime, Shanghai, China). Equal amounts of protein (20–40 μg) were resolved by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently transferred onto polyvinylidene difluoride (PVDF) membranes (Beyotime, Shanghai, China). Membranes were blocked with 5% non-fat dry milk or bovine serum albumin (BSA) in Tris-buffered saline containing 0.1% Tween-20 (TBST) for 1 h at room temperature. Membranes were then incubated overnight at 4 °C with primary antibodies diluted in blocking buffer. The primary antibodies used were: anti-PRMT1 (1:1000; ab190892; Abcam), anti-E-cadherin (1:1000; ab40772; Abcam), anti-N-cadherin (1:5000; ab76011; Abcam), anti-Vimentin (1:1000; ab92547; Abcam), anti-ADMA (Asymmetric dimethylarginine; 1:100; ab413; Abcam), anti-phospho-STAT3 (Tyr705) (p-STAT3; 1:2000; ab76315; Abcam), anti-Lamin B (1 µg/mL; ab232731; Abcam), anti-VEGFA (1 µg/mL; ab46154; Abcam), anti-IL-6 (1:1000; ab9324; Abcam), anti-c-myc (1:1000; ab32072; Abcam), anti-GPX4 (concentration not specified in draft, standard dilutions 1:1000), anti-GAPDH (concentration not specified, 1:5000–1:10,000), and anti-β-actin (1 µg/mL; ab8226; Abcam). After washing with TBST, membranes were incubated with appropriate HRP-conjugated secondary antibodies (e.g., ab7090, 1:2000; Abcam) for 2 h at room temperature. Protein bands were visualized using an enhanced chemiluminescence (ECL) detection kit (Thermo Fisher Scientific, Inc., Waltham, MA, USA) and imaged using a suitable detection system. Band intensities were quantified using ImageJ software (NIH, Bethesda, MD, USA) or similar software, with GAPDH or β-actin serving as loading controls.

Cell viability assay (CCK-8)

Cell viability was assessed using the Cell Counting Kit-8 (CCK-8; Dojindo Laboratories, Kumamoto, Japan). HN6 and SCC25 cells were seeded into 96-well plates at a density of 1000 cells per well. After adherence and appropriate treatments (e.g., varying concentrations of Doxorubicin (DOX) or Cisplatin (CDDP)), 10 μL of CCK-8 solution was added to each well, followed by incubation for 2 h at 37 °C. The absorbance at 450 nm was measured using a microplate spectrophotometer (Thermo Fisher Scientific). Dose–response curves were fitted using a four-parameter logistic regression (4PL) model in GraphPad Prism 9.0, and half-maximal inhibitory concentrations (IC50 values, mean ± SD) were calculated from 3 independent experiments. Statistical significance between IC50 values of sh-NC and sh-PRMT1 groups was assessed using an extra-sum-of-squares F-test.

Colony-formation assay

Cells were trypsinized to single-cell suspensions and seeded in 6-well plates (500–1,000 cells/well) in complete medium. After 10–14 days (with medium changes every 3–4 days), colonies were fixed (4% paraformaldehyde, 15 min) and stained with 0.5% crystal violet (30 min), rinsed, air-dried, and colonies ≥ 50 cells were counted by two blinded observers. For quantification, colony counts per well were averaged across three biological replicates and analyzed by two-sided Student’s t-test (Supplementary Fig. S1c).

Cell proliferation assay (BrdU and EdU)

Cell proliferation was evaluated using Bromodeoxyuridine (BrdU) incorporation. Cells were incubated with BrdU labeling solution for a specified period, 3 h. Subsequently, cells were fixed, permeabilized, and treated with DNase to expose incorporated BrdU. Detection was performed using an anti-BrdU antibody conjugated to a fluorophore, followed by counterstaining with DAPI (4′,6-diamidino-2-phenylindole) to visualize nuclei. Images were acquired using a fluorescence microscope (Leica, Hilden, Germany), and the percentage of BrdU-positive cells relative to the total number of DAPI-stained cells was calculated.

For experiments related to ferroptosis rescue, cell proliferation was assessed using the Cell-Light EdU DNA Cell Proliferation Kit (RiboBio, Guangzhou, China). Briefly, HN6 and SCC25 cells were incubated with 50 μM EdU solution for 2 h. Cells were then fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X-100. EdU incorporation was detected by click chemistry using an Apollo dye solution according to the manufacturer’s protocol. Nuclei were counterstained with DAPI. EdU-positive cells were visualized and quantified using fluorescence microscopy (Leica, Hilden, Germany).

Transwell invasion assay

Cell invasion capacity was measured using Transwell chambers (8 μm pore size; Corning, NY, USA) coated with Matrigel (BD Biosciences, Franklin Lakes, NJ, USA). HN6 or SCC25 cells (approximately 5 × 10⁴ to 1 × 10⁵ cells) were resuspended in 200 μL of serum-free DMEM and seeded into the upper chamber. The lower chamber was filled with 600 μL of DMEM containing 20% FBS as a chemoattractant. After incubation for 48 h at 37 °C, non-invading cells on the upper surface of the membrane were removed with a cotton swab. Cells that had invaded through the Matrigel and membrane to the lower surface were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet solution. Invaded cells were photographed and counted in several randomly selected fields under a microscope (Olympus Optical Co., Ltd., Tokyo, Japan).

In Vivo xenograft tumor model

All animal experiments were approved by the Animal Ethics Committee of Beijing Viewsolid Biotechnology Co. LTD. All animal experiments were conducted in accordance with the ARRIVE guidelines. Male BALB/c nude mice (4–6 weeks old) were obtained from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). For subcutaneous xenografts, 1 × 10⁶ transfected HN6 cells (e.g., sh-NC, sh-PRMT1, STAT3-WT, STAT3-KO) suspended in 100 μL PBS were injected into the right flank of each mouse. Tumor growth was monitored regularly by measuring tumor dimensions with calipers. Tumor volume was calculated using the formula: Volume = (length × width²) / 2. For therapeutic studies, treatments began when tumors reached a palpable size. Treatment groups received intraperitoneal injections of Cisplatin (CDDP; 150 mg/kg, administered twice a week), anti-PD-1 antibody (200 μg per mouse, frequency specified if different), or MS023 (PRMT1 inhibitor; 80 mg/kg, intraperitoneal injection, frequency specified if different), or saline control. After 28 days (or as specified), mice were euthanized by cervical dislocation. Tumors were excised, weighed, photographed, and processed for histological or molecular analysis. For metastasis studies, lung tissues were also collected at necropsy.

Lung metastasis assessment and hematoxylin–eosin (HE) staining

Harvested lung tissues were fixed in 4% paraformaldehyde and embedded in paraffin. Serial Sects. (4 μm thickness) were prepared and stained with Hematoxylin and Eosin (HE) following standard protocols. The number of visible metastatic nodules on the lung surface was counted macroscopically before fixation, and/or microscopically on HE-stained sections. Representative images of lung histology were captured using a microscope (Olympus, Japan).

Co-immunoprecipitation (Co-IP)

To investigate protein interactions, Co-IP assays were performed using a Co-IP kit (Abison Biotechnology Co. Ltd, China). HN6 or SCC25 cells were treated with the proteasome inhibitor MG-132 (10 µM) for several hours before lysis to stabilize protein complexes. Cells were lysed in immunoprecipitation buffer. Cell lysates were clarified by centrifugation, and the supernatants were incubated overnight at 4 °C with primary antibodies against the target protein (e.g., anti-STAT3 for detecting interaction with ADMA-modified proteins, or anti-HA/Flag for tagged proteins, or anti-PRMT1) or control IgG. Protein A/G agarose beads were added and incubated for another 2–4 h to capture antibody-protein complexes. Beads were extensively washed with lysis buffer. Immunoprecipitated proteins were eluted by boiling in SDS loading buffer and analyzed by Western blotting using antibodies against the potential interaction partners (e.g., anti-ADMA, anti-STAT3, anti-PRMT1, anti-HA, anti-Flag). Input lysates were simultaneously analyzed to confirm the expression of target protein.

Flow cytometry analysis of immune cells and ROS

For analysis of immune cell in peripheral blood from treated mice, 100 μL of whole peripheral blood was collected. Red blood cells were lysed, and remaining leukocytes were stained with fluorochrome-conjugated monoclonal antibodies against surface markers: anti-CD3 (ab38483; Abcam), anti-CD8 (ab38483; Abcam). For intracellular staining, cell were fixed, permeabilized, and stained with an antibody against Granzyme B (GranB; ab38483; Abcam). Staining was performed in the dark for 30 min at 4 °C. After washing, cells were analyzed using a flow cytometer to quantify the percentages of CD3 + CD8 + T cells and GranB + cells within the CD8 + population. For the detection of intracellular Reactive Oxygen Species (ROS), cultured cells were incubated with the ROS-sensitive fluorescent probe DCFH-DA (included in kit S0033S, Beyotime, Shanghai, China) according to the manufacturer’s instructions. After incubation, cells were washed, harvested, and resuspended in PBS. Fluorescence intensity, which correlates with ROS levels, was measured using flow cytometry. Data analysis included measuring the mean fluorescence intensity (MFI).

Chromatin immunoprecipitation (ChIP) assay

ChIP assays were conducted using a commercial kit (Beyotime, Beijing, China) following the manufacturer’s instructions. Briefly, 293 T cells (or OSCC cells if applicable) were cross-linked with 1% formaldehyde for 10 min at room temperature. Cross-linking was quenched with glycine. Cells were lysed, and chromatin was sheared into fragments of approximately 200–500 base pairs using sonication. The sheared chromatin was pre-cleared and then incubated overnight at 4 °C with an antibody against STAT3 or a control IgG antibody. Immune complexes were captured using Protein A/G agarose beads. After extensive washing, cross-links were reversed, and DNA was purified. The enrichment of specific GPX4 promoter regions (P1, P2, P3, and P4, defined by primer pairs) in the immunoprecipitated DNA was quantified by quantitative polymerase chain reaction (qPCR) using specific primers for these regions. Results were normalized to input DNA.

Luciferase reporter assay

A putative promoter region of the human GPX4 gene containing predicted STAT3 binding sites (WT2 region: -1136 to -491), alongside a version with the binding sites mutated (MUT2), were cloned into the pGL3-Basic luciferase reporter vector (Promega, Madison, WI, USA). The empty pGL3-Basic vector served as a negative control. 293 T cells were co-transfected with one of these GPX4 promoter-luciferase constructs along with a STAT3 expression vector (or an empty vector control) and a Renilla luciferase vector (pRL-TK) for normalization. At 48 h post-transfection, cells were lysed, and luciferase activities (Firefly and Renilla) were measured using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s protocol. Firefly luciferase activity was normalized to Renilla luciferase activity to control for transfection efficiency.

Measurement of ferrous iron (Fe2 +) levels

Intracellular labile ferrous iron (Fe2 +) levels were quantified using a commercial Iron Assay Kit (ab83366; Abcam, Shanghai, China) according to the manufacturer’s instructions. Briefly, cell lysates were prepared, and the assay utilizes a chromogen that reacts specifically with Fe2 + to produce a colored product. The absorbance was measured at the recommended wavelength using a microplate reader, and Fe2 + concentrations were calculated based on a standard curve.

Measurement of lipid peroxidation (MDA) and glutathione (GSH)

Levels of malondialdehyde (MDA), an indicator of lipid peroxidation, and glutathione (GSH), a key antioxidant, were measured in cell lysates or tissue homogenates using commercial enzyme-linked immunosorbent assay (ELISA) kits. MDA levels were determined using the MDA assay kit (ab118970; Abcam, Shanghai, China), and GSH levels were measured using the GSH assay kit (ab65322; Abcam, Shanghai, China), strictly following the protocols provided by the manufacturer. Absorbance readings were taken using a microplate reader, and concentrations were determined by comparison to standard curves provided with the kits.

Statistical analysis

All quantitative data are presented as the mean ± standard deviation (SD) from at least three independent experiments or biological replicates. Statistical analyses were performed using GraphPad Prism Software (version 9, GraphPad Software, La Jolla, CA, USA). Comparisons between two groups were made using a two-tailed Student’s t-test. Comparisons among three or more groups were performed using one-way analysis of variance (ANOVA) followed by an appropriate post-hoc test (e.g., Tukey’s test). Correlations were assessed using Pearson’s correlation coefficient. Kaplan–Meier survival curves were compared using the log-rank test. A P-value less than 0.05 (P < 0.05) was considered statistically significant, with specific P-values provided in the figures.

By William Schomberg and David Milliken

LONDON (Reuters) -When accountants at mid-tier firm Moore Kingston Smith began using artificial intelligence to speed up their work, profit margins jumped.

Colleagues in another team running checks against corporate fraud created a report for customers in two hours, something that previously took two weeks.

The rollout of AI is raising hopes that Britain’s economy can escape the productivity problem that has dogged it for two decades, even as slow growth pushes finance minister Rachel Reeves towards tax hikes in Wednesday’s budget.

Economists say the dominance of ​services businesses in Britain’s private sector compared to other countries could mean higher rewards if it rapidly adopts AI in powerhouse sectors such as accountancy and finance.

Ratings agency Moody’s said on Friday the UK could gain more than other countries from the advances in the technology.

Becky Shields, MKS’ head of ‌digital transformation, said AI was freeing staff from repetitive work and giving them more time to work with clients.

“The large language models that underpin all of this technology are evolving all the time. They’re getting better and better with every iteration,” she said.

UK PLC IS AI-READY

Services make up 80% of Britain’s economy, the same as the United States – and account for a bigger ‌share once services generally provided by the state are stripped out.

MKS, with about 1,500 UK staff, is applying its platform based on Google’s Gemini 2.5 model to a growing range of work. Shields said it was still a learning process, but its positive impact was clear.

A team that used AI four times more intensively than another group reported a profit margin 8 percentage points higher, she said.

Rather than ask for proof of orders, invoices, bank statements and other documents for a sample of transactions, the team using AI let clients upload entire datasets which MKS was able to analyse automatically.

Initial extra work is now paying off as the process can be applied more widely. The reduced paperwork helps clients who say they chose the firm for its AI adoption, Shields said.

“You can do a lot with a little with how the technology currently sits,” she added, describing the cost of AI as “pennies in the pound” compared with other technology.

For Britain’s economy – and struggling Prime Minister Keir Starmer – improving productivity ⁠is a major challenge.

An expected downgrade by budget forecasters of the economy’s underlying growth potential, reflecting past disappointments,‌ is set to blow a hole in the public finances, meaning Reeves is likely to increase taxes on Wednesday to reassure nervous bond investors that she can cut borrowing.

PRODUCTIVITY DRAGS

As elsewhere, improvements in Britain’s productivity slowed after the 2007-08 global financial crisis, leading to almost 20 years of weak growth and frustration among voters.

Feeble productivity gains account for half of Britain’s slowdown in pay growth since 2008, according to the National Institute of Economic and Social Research, a think tank.

Starmer’s government ‍is trying to streamline the planning system, modernise infrastructure and improve skills to raise productivity and get the economy firing again. It also hopes that AI can inject more efficiency into public services.

“Productivity isn’t everything, but in the long run it is almost everything,” Nobel Prize-winning economist Paul Krugman said in 1990. Bart van Ark, head of the University of Manchester’s Productivity Institute, said that for Britain’s government “almost everything in the short run is productivity”.

Britain has the highest inflation rate among the Group of Seven rich nations and too many people dropping out of the jobs market. Its business investment rate was the G7’s second-lowest in 2024, although comparable to that for the United States, which ​manages far better productivity.

But AI could be a card up Britain’s sleeve.

One of Reeves’ deputy ministers, Torsten Bell, who is helping to write her budget, argued in his 2024 book that Britain “can and should ride the services wave”.

Bolstered by its expertise in financial and business services, law, ‌education and architecture, Britain had a services trade surplus of $248 billion in 2024, second only to the United States, World Trade Organization figures show.

British services exports accounted for just over 7% of the world’s total, again the second-highest after the U.S.

Less clear is the AI upside for British factories. They are struggling in the face of high energy, labour and raw materials costs, low public infrastructure investment and shifts in trading rules.

Flooring materials maker Amtico, in England’s Midlands region, uses AI to plan production. But its next big investment decision is about expanding the use of robotics to offset some of Britain’s high costs for manufacturers.

“I am looking to take my most labour-intensive processes and invest my way out of it,” Jonathan Duck, Amtico’s chief executive, said.

Many firms are still smarting from a labour tax hike in Reeves’ first budget last year. Employers say Reeves would risk her growth agenda if she increases tax on them again.

GROWTH UP, BUT JOBS DOWN?

Analysts say it is too early to be sure about AI’s longer-term impact on economic growth as estimates range widely given the uncertainties about its real-world applications, but most agree it will not be immediate.

Bank of England Governor Andrew Bailey, who sees ⁠AI as a potential game-changer, said last month it took around 40 years between Thomas Edison first wiring up a light bulb and the impact of electricity showing up in the productivity ​statistics.

“We are still at the experimentation stage with AI, so investment and persistence is crucial,” Bailey said.

The University of Manchester’s Van Ark sees AI adding 0.1 to 0.2 percentage ​points to annual growth in the coming years. That would help an economy growing by about 1.5% a year but will not spare the current government from tough budget choices.

Paul Dales, chief UK economist at Capital Economics, said AI was likely to speed up growth in the mid-2030s, with Britain seeing more adoption than Europe’s other big economies due to its more hands-off approach to regulation and labour laws.

Andrew Wishart, an economist at bank Berenberg, said improved productivity in higher-value corporate sectors offered signs that a ‍broader change was already underway.

“If we don’t have a sharp rise in taxes, ⁠I think we should see it in business earnings,” he said.

Risks from the rise of AI include the possibility that its gains flow mostly to bigger firms with more funds to invest, potentially leading to a less competitive economy and aggravating Britain’s geographic imbalances.

Some firms in highly regulated sectors, such as accountancy, worry that rule-makers will not keep pace with the technology.

“The challenge for businesses is not getting a sense right now of what is and isn’t allowed,” Esther Mallowah, head of tech policy at the Institute of Chartered Accountants in England and Wales, said.

Another big risk from ⁠a more automated future, its impact on the labour market, is becoming a little clearer.

A survey by the Chartered Institute of Personnel and Development this month showed 17% of private sector employers expected to reduce headcount over the next 12 months as a result of AI. Only 6% planned an increase.

At MKS, the number of graduates hired this year was cut. But Shields said the ‌reduction was intended as a “short shock” to speed up the way staff adapt to AI.

Hiring would probably return to normal after a year to ensure the human touch is not lost.

“Our clients trust us to do more with their business whether it’s wanting more assistance or ‌new services,” Shields said. “There is no expectation that will be a long-term trend.”

(Writing by William Schomberg; Editing by Catherine Evans)

By Jason Chau

  JD.com's supply-chain technology unit has started gauging investor interest for its long-awaited Hong Kong initial public offering, amid a fundraising boom in the city's financial markets. 

  In an exchange filing posted on Sunday, Jingdong Industrials said it intends to use the net proceeds raised to further enhance its industrial supply-chain capabilities, expand internationally and pursue potential strategic investments or acquisitions. 

  If completed, the listing would conclude a two-year IPO process. 

  BofA Securities, Goldman Sachs and UBS are among the banks advising on the potential listing. 

  JD.com first disclosed plans to list Jingdong Industrials in March 2023 alongside a listing for property unit Jingdong Property. Jingdong Industrials' listing application was approved by China's securities regulator in September this year. 

  Chinese e-commerce giant JD.com has previously spun off businesses through listings over the years, including online healthcare unit JD Health International and supply-chain solutions provider JD Logistics. 

  JD.com's Hong Kong-listed shares have slid nearly 18% this year, despite the benchmark Hang Seng Index rising close to 28% as investors returned to Hong Kong equities on renewed confidence in China's technology-sector growth. 

  Write to Jason Chau at jason.chau@wsj.com 

  (END) Dow Jones Newswires

  November 24, 2025 01:07 ET (06:07 GMT)
Copyright (c) 2025 Dow Jones & Company, Inc.