Category: 3. Business

The proposed smart platform was implemented in a controlled greenhouse environment. The sensor network successfully collected real-time data on soil moisture, temperature, and humidity. ESP32 transmitted this data wirelessly to the ThingsBoard platform, where it was visualized on a user-friendly dashboard shown in Fig. 9. The Raspberry Pi captured images of tomatoes at regular intervals. The YOLOv8 model accurately detected tomatoes in the images and classified them according to their ripening stage (green, half-ripened, fully ripened). The processed image data, including bounding boxes and ripening stage labels, was transmitted to the ThingsBoard platform.

Image with ripening stage detection: Convolution neural network model

The YOLOv8 model accurately detected the ripening stages of the tomatoes. The Raspberry Pi captured images of the tomatoes at different stages of ripeness, and the model classified them into green, half-ripened, and fully ripened categories with high accuracy. Figure 10 presents an example image captured by the Raspberry Pi camera. The image overlays bounding boxes around detected tomatoes with labels indicating their respective ripening stages (green, half-ripened, or fully ripened) predicted by the YOLOv8 model. This visual representation provides valuable information for harvest planning and resource allocation shown in Figs. 11 and 12.

Confusion matrix

To examine the effectiveness of the YOLOv8n model, a confusion matrix was created (Figs. 17, 18, 19). The matrix displays the predicted versus actual classes, allowing us to assess the model’s accuracy for each class. The results show that the model performs well in distinguishing between different ripeness stages. For instance, ‘l_green’ instances are mostly correctly classified, while some misclassifications occur between ‘b_half_ripened’ and ‘b_green’. The confusion matrix reveals that the ‘l_green’ class has the highest number of correct predictions, with a few instances of misclassification in other categories. This indicates that the model can accurately identify green tomatoes, but there is some room for improvement in distinguishing between half-ripened and fully ripened stages.

Model performance

Detection accuracy

The model demonstrates strong object detection capability across all six classes, with the highest accuracy in detecting l_green and l_fully_ripened tomatoes. Detection performance is slightly lower for b_fully_ripened and b_half_ripened, which can be attributed to the lower number of annotated samples.

In Table 4, the precision, recall, f1-score, and support columns show the results. Precision indicates the accuracy of the positive predictions made by the model, recall indicates how well the model can identify the overall positive class, and F1-score indicates the accuracy of the model. Based on the evaluation matrix results from Fig. 13, the model has an accuracy of 52% in testing the test data.

Visual prediction result

• The output detection images in figure clearly show accurate bounding boxes and class labels overlaid on test images, confirming the model’s robustness under varied lighting and angle conditions.

• Even in dense clusters or partial occlusions, the YOLOv8n model maintained reliable detections, making it viable for real-time agricultural applications.

Cloud dashboard: visualization and analysis

Figures 14 and 15 depicts the ThingsBoard dashboard displaying real-time sensor data from the greenhouse. The dashboard includes widgets showcasing the current values of soil moisture, temperature, and humidity. Users can customize the dashboard to display historical data in the form of graphs and charts, providing insights into trends and fluctuations over time. Analysing these trends enables farmers to identify potential issues like rising temperatures or declining soil moisture levels and take corrective actions before they negatively impact crop health.

Data visualization

Figure 16 illustrates a sample graph generated from the sensor data collected by the Thingsboard platform. The graph depicts the variations in temperature within the greenhouse over a specific period. The cloud-based dashboard provided a comprehensive view of the sensor data and ripening stages. The dashboard displayed real-time graphs of soil moisture, temperature, and humidity levels, as well as images of the tomatoes with their respective ripening stage.

Analysis

In this section, we present the results and analysis of the YOLOv8 model trained to detect the ripening stages of tomatoes. Various visualizations and metrics have been used to evaluate the model’s performance.

The instance distribution graph (Fig. 17) shows the number of instances for each class (‘b_fully_ripened’, ‘b_half_ripened, b_green’, ‘l_fully_ripened’, ‘l_half_ripened’, ‘l_green’). It highlights that the ‘l_green’ class has the highest number of instances, followed by ‘b_green’. The bounding box distributions (Fig. 18) display the normalized x and y coordinates, showing the density and distribution of bounding boxes in the images. The width and height distributions (Fig. 19) indicate the range of sizes of the bounding boxes, revealing a correlation between width and height.

Figure 20 presents the training and validation loss curves over 20 epochs. The training and validation box loss, class loss, and DFL loss are plotted. The results indicate a steady decrease in losses, suggesting that the model is learning effectively. Metrics such as precision, recall, mAP50, and mAP50-95 are also displayed, showing improvements as training progress. These metrics are critical for evaluating the model’s performance in detecting and classifying tomato ripeness stages.

Energy consumption

Understanding the energy consumption of hardware components is critical for designing efficient and sustainable systems, especially in applications like smart greenhouses where devices are continuously operational. This section presents a detailed analysis of the Raspberry Pi 3 Model B and the ESP32 microcontroller, two essential parts of the smart greenhouse system, and how much energy they require.

Energy consumption of ESP32⁸

The ESP32 microcontroller is designed for low-power applications and offers significant energy efficiency. It operates at a voltage of 3.3 V and has distinct power consumption modes depending on its activity:

Active mode

When actively processing data or transmitting information, the ESP32 consumes approximately 160 mA.

The power consumption in this mode is calculated as

Sleep mode

In deep sleep mode, the ESP32 significantly reduces its power consumption to about 10 µA. The power consumption during sleep is:

Assuming the ESP32 is active for 12 h a day and in deep sleep for the remaining 12 h, the daily energy consumption is:

Total daily energy consumption

Wi-Fi transmission power

Wi-Fi communication is a major factor affecting ESP32’s power usage. The ESP32 uses ~ 260 mA when transmitting data over Wi-Fi, significantly increasing its power draw. To estimate the additional energy consumption:

Wi-Fi Usage Scenario: Assuming the ESP32 transmits data for 15 min per hour throughout a 12-hour active period:

Revised Total ESP32 Energy Consumption:

Energy consumption of raspberry Pi 3 B⁷

The Raspberry Pi 3 Model B is intended for more demanding computational operations and runs at a higher voltage of 5 V. The way it operates affects how much power it uses:

Active mode

When fully operational, the Raspberry Pi consumes approximately 500 mA. The power consumption is:

Idle mode

When in idle mode, the Raspberry Pi’s consumption drops to about 400 mA. The power consumption is:

For a continuous operational period of 24 h, the daily energy consumption is:

Wi-Fi power consumption

Raspberry Pi Wi-Fi transmission draws ~ 300 mA additional current, leading to increased consumption.

Total energy consumption

The energy consumption comparison between the ESP32 and the Raspberry Pi reveals substantial differences as shown in Fig. 21.

ESP32

Consumes about 8.91 Wh/day when active for 12 h and in sleep mode for 12 h. This low energy requirement highlights its suitability for battery-powered and energy-efficient applications.

Raspberry Pi

Consumes approximately 78 Wh/day when continuously active. This higher energy consumption reflects its more intensive computational capabilities and constant operation.

Figure 21 shows the comparative energy usage, highlighting the need for power-efficient alternatives in future deployments (e.g., LoRa, Edge TPU).

The proposed work considered a 9 V Hi-Watt battery as a potential power source. While we have not physically tested this battery, analyzed its theoretical performance based on known specifications and expected power consumption patterns. The Hi-Watt 9 V battery is a commonly available dry-cell battery with an estimated 500mAh capacity, providing approximately 4.5Wh of total energy.

Since the ESP32 operates at 3.3 V, a voltage regulation circuit (such as an LDO or buck converter) would be required to step down the voltage. This introduces additional power losses, reducing the effective energy available to ESP32. Moreover, battery performance is influenced by nonlinear discharge characteristics, meaning that as the battery depletes, its voltage output gradually drops, affecting system stability.

The analysis highlights the importance of selecting a battery based on actual usage requirements rather than theoretical capacity alone. Factors like duty cycle, voltage regulation losses, and environmental conditions can significantly impact battery life. If a longer operational period is required, alternative power solutions such as higher-capacity Li-ion or Li-Po batteries, energy harvesting modules, or solar-powered setups could be considered.

The battery depletion curve for the ESP32 shown in Fig. 22 is not linear due to several factors, including internal resistance, chemical reaction kinetics, and discharge rate effects. As seen in the graph, the depletion follows an exponential decay trend rather than a straight-line decline. Initially, energy drains more rapidly due to higher available charge, but as the battery nears depletion, the rate of voltage drop slows down.

The ESP32’s energy consumption varies depending on its operating mode. In active transmission states (Wi-Fi/Bluetooth operations), power draw is significantly higher, whereas in deep sleep modes, the current consumption is minimal. The graph reflects an estimated depletion time of around 17–18 h, assuming a mixed operation profile. Once the battery level reaches the low battery threshold (20%), the system may enter power-saving modes or shut down due to insufficient voltage supply.

Understanding throughput vs. delay

Throughput refers to the rate at which data is successfully transmitted and received over the network, typically measured in kilobits per second (Kbps) or megabits per second (Mbps). Delay, on the other hand, represents the latency or time taken for data to reach the destination, which includes processing, queuing, and transmission delays.

In an ideal scenario, high throughput and low delay are desirable for efficient data transmission. However, real-world environments introduce factors such as:

• Network Congestion: High data traffic may lead to increased queuing delays.

• Wi-Fi Signal Strength: Weak signal strength can reduce data rates and increase retransmissions.

• Cloud Processing Time: The ThingsBoard platform introduces an additional delay due to server-side data processing and dashboard updates.

From the Throughput vs. Delay graph shown in Fig. 23, it is evident that as throughput increases, delays tend to rise beyond a certain threshold. This can be attributed to limited bandwidth and network contention, where a higher data rate results in increased packet queuing and retransmissions. The ESP32, operating on a typical 2.4 GHz Wi-Fi connection, experiences latencies ranging from a few milliseconds to several hundred milliseconds, depending on environmental interference and cloud response times.

Spotify Technology SA a Luxembourg-based company, which offers digital music-streaming services. The Company enables users to discover new releases, which includes the latest singles and albums; playlists, which includes ready-made playlists put together by music fans and experts, and over millions of songs so that users can play their favorites, discover new tracks and build a personalized collection. Its users can either select Spotify Free, which includes only shuffle play or Spotify Premium, which encompasses a range of features, such as shuffle play, advertisement free, unlimited skips, listen offline, play any track and audio. The Company operates through a number of subsidiaries, including Spotify LTD and is present in over 20 countries. Its service offers a music listening experience without commercial breaks.

Our study reported a high rate of patients developing new pituitary deficiencies, up to 37.1% of 70 patients with a median follow-up of 20.7 months. These deficiencies could occur within the first six months following the end of radiotherapy. Few prospective studies have focused on pituitary deficiencies after irradiation and all of them used photons. Despite a short follow-up compared to retrospective studies, our cohort is one of the largest among prospective studies and the only one including multiple histologies. Similar to other studies, we report a high frequency of HP deficiencies [1, 9].

Median age was 60 years in our cohort as opposed to 47 years on average in other studies. Women were more frequent than men, which was related to predominance of meningiomas in this series. Patients irradiated for nasopharyngeal carcinoma (NPC) had the highest risk with a mean prevalence of 0.68 [1]. Dose constraints to the pituitary gland and hypothalamus were left to physician’s appreciation. In monitored patients, most pituitary glands received a mean dose > 50 Gy which prevented identification of deterministic effects and thresholds. Indeed, a mean PG dose > 50 Gy has been identified as a risk factor for pituitary deficiency in multivariate analysis [10]. Similarly, a correlation between the dose to the pituitary and the occurrence of hormonal axis insufficiency could not be observed [1], possibly because testing was primarily conducted in at-risk patients. In this population of patients receiving > 50 Gy to their PG per selection on endocrine monitoring, mean HT dose was not a risk factor (p = 0.56) although Partoune et al.reported a risk progressively increasing without a clear threshold [11]. It has been described that the higher the dose to the pituitary, the shorter the latency of GH deficiency [7]. This effect has never been reported for the other axes. However, it could be envisaged that it exists to a lesser extent due to a lower radiosensitivity. Larger prospective studies with a wider range of doses and regular endocrine assessments would be necessary to clarify its importance. Kyriakakis et al. described dose thresholds for every axis, from 10 Gy for GH to 50 Gy for TSH [12] and Darzy et al. reported mostly isolated GH deficiencies below 30 Gy [7].

Our data of all five axes show that 37.1% of patients developed any new pituitary dysfunction within a median time of 14.1 months, including 17.1% within the first 6 months. Endocrine deficiencies have been reported as early as 3 months after radiotherapy, even at doses below 30 Gy [13]. We showed that pituitary deficiency occurred in at least one hormonal axis in 28.6% of cases and the number of surgical procedures before radiotherapy was an important risk factor; suggesting that it is critical to assess follow up levels with respect to baseline levels. Interestingly, among the 20 patients with deficits before radiotherapy, there was a trend toward cumulative or progressive pituitary dysfunction in patients with pre-existing impairment, consistent with a possible additive effect of surgery and irradiation. Moreover, repeated surgical interventions may increase the risk of damage to the pituitary gland or stalk, compromise vascular supply, or lead to scarring and adhesions that exacerbate radiation effects. This cumulative burden may explain the higher frequency of new deficits observed in patients with multiple surgeries [14]. Finally, GH dysfunction was 9.7% within 27.1 months, lactotroph dysfunction 35% within 13.5 months, gonadotroph 23.2% within 24.5 months, ACTH 3.2% with 27.0 months and TSH 7.8% within 21.8 months. In a recent meta-analysis, mean prevalence of endocrine insufficiency of 1 axis was 19%, 2 axes 22%, 3 axes 5% and panhypopituitarism 17% [1], of overall similar magnitude as in our cohort. Mean prevalence of GH deficiency was 40%, of higher magnitude than in our series, prolactin 22% of slightly lower magnitude, gonadotropin 20% of similar magnitude, while ACTH 16% and TSH 16% of higher magnitude than in our series [1]. A significant correlation is indeed observed between any endocrine insufficiency and follow-up time [1] and could explain the slight differences. Pituitary axes show different radiosensitivity in our series and the literature [1] and long term data are warranted for more personalized replacement therapies [7, 14, 15] and to promote long-term follow-up as HP deficiencies frequently occur after some years [10]. Identified differences in sensitivity vary between the anterior pituitary axes, with the growth hormone axis being the most easily damaged by irradiation and the thyroid hormone axis the least sensitive. Pituitary gland protection and early detection of deficiencies need further investigations. Of note, the hypothalamus seemed to be more vulnerable to radiation dose compared to the pituitary gland, which warrants further systematic, standardized and long-term monitoring data to establish reliable normal tissue complication probability (NTCP) models for HP deficiencies [1].

The rationale for using protons over photons lies in their particular dose deposition. The superiority of proton therapy lies in their excellent distal dose fall off at the end of the Bragg peak with virtually no dose behind, and this is particularly interesting to spare substantial volumes of distant organs. However, for relatively superficial tumors on the CNS and HN, most commercial pencil beam scanning proton therapy machines use a minimal 100 MeV, which requires the use of a range shifter to reach tumors less than 7.5 cm below the skin. Such energy degraders widen the lateral penumbra [16]. In tumors abutting the pituitary gland or the optic chiasm, stereotactic irradiation was sometimes used as the boost component to yield the steepest gradient-generating technique. Proton therapy however significantly spares more distant organs and the brain itself, which can have significant advantages on cognition compared to photons. Hypothalamic-pituitary deficiencies after proton therapy were described in two papers [17, 18]. In these retrospective studies of 74 (Lamba et al.) and 103 patients (De Marzi et al.), patients were treated for meningioma, chordoma or chondrosarcoma with a median mean dose to pituitary gland of 51.4 Gy and 54 Gy. Lamba et al. reported a new deficiency rate of 20% for any axis, occurring at a median time of 0.9 to 2.7 years after radiation depending on the axis. No difference was found between protons and photons regarding pituitary deficiencies. De Marzi et al. reported a 44% rate for any axis but no information regarding the duration of the follow-up was available. We conducted a comprehensive analysis of endocrine deficiencies and found that new deficits occurred in 40% of patients undergoing radiotherapy within a median follow-up of 5.6 years [19], in line with a more recent study [1] and the time-effect relationship. We observed similar rates of pituitary dysfunction between patients treated with protons alone and those who received combined photon-proton therapy. This may reflect comparable radiation doses delivered to the hypothalamo-pituitary axis across modalities, particularly in anatomically complex cases where sparing was limited regardless of technique. Few studies reported hypothalamic-pituitary deficiencies after proton therapy but no difference was found between proton-based and photon-based irradiations of tumors close to the PG, suggesting similarly steep gradient gradients.

A limitation of our study could be the absence of systematic dynamic tests, which probably led to a significant underestimation of the rate of somatotrophic and corticotropic partial deficiencies. However, a recent study of 246 patients referred to endocrinologists after cranial radiotherapy showed ACTH deficit was rare, and never isolated. The authors suggested that it may not be necessary to carry out a dynamic test for ACTH if no other deficits are diagnosed [15]. Another issue is GH supplementation in adults, especially in the context of a tumor. It remains widely debated, which might be one of the reasons why these deficiencies are not optimally investigated in routine practice [20]. GH replacement seemed to improve well-being parameters in adults but there were safety concerns [20, 21] about whether GH replacement increases future cancer risk. It is difficult to identify factors that modulate cancer risk in adults and there is some evidence for in-vitro pro-neoplastic properties and increased serum concentrations of IGF-I that could independently contribute to worse tumor or mortality outcomes in at-risk populations. Similarly, radiation-induced gonadotropic deficits may be overestimated in the absence of correction of associated asymptomatic hyperprolactinemia. We found that BMI < 25 kg/m² is associated with earlier pituitary deficiency. To our knowledge, this association has never been reported and currently lacks pathophysiological explanation. The interpretation of prolactinemia may be particularly complex in this oncology context. This can only emphasize the need for more systematic and long-term assessment of the HP axes in oncology adults [22]. However, we showed a relative lack of patient adherence to follow-up of long-term side effects, particularly if asymptomatic. A limitation of our study and many others lies in a selection bias toward monitoring endocrine deficiencies only in patients at high risk of deficiency. Some patients were lost to follow-up at our facility or did not want to continue endocrine monitoring. At 3 years, 14 patients were evaluable and achieved over 85% compliance. This is common in oncology and might be improved with better education of patients, radiation oncologists and family doctors to promote a long-term standardized follow-up of those patients, considering that blood testing is hardly invasive. To overcome sampling and quantification biases, we considered using the raw data of plasma hormone levels from the cohort to describe the effects of irradiation. This approach provides additional information and has never been described in the context of hypothalamic-pituitary deficiencies. Moreover, it could be a first step towards modeling radiation-induced hypothalamic-pituitary effects. We included patients with pituitary resection and pre-existing pituitary deficits in the analysis, as they are at high risk of further deterioration, as long as they did not have complete panhypopituitarism. Monitoring this subgroup provides valuable insight into the potential additive or progressive impact of radiotherapy on residual pituitary function, which reflects real-world clinical scenarios and underscores the importance of individualized long-term endocrine follow-up.

The fact that this series is performed in patients undergoing proton therapy should not be misunderstood as an investigation of proton-specific side effects. More than ever, comparative studies and randomized photons/protons trials are warranted. Additionally, endocrine management guidelines [8] customized to the adult oncology context would be helpful to better standardize endocrine monitoring, its duration, based on the tumor characteristics and dose level, and replacement therapies in routine care [11, 12].

I think these UK income stocks, exchange-traded funds (ETFs), and investment trusts are worth serious consideration from serious dividend investors. Here’s why.

Today Phoenix Group (LSE:PHNX) has the fourth-highest dividend yield on the FTSE 100. At 7.9%, its yield is only topped by those of Taylor Wimpey, WPP, and Legal & General.

But unlike those first two blue chips, Phoenix’s yield hasn’t been supercharged by extreme share price weakness. The financial services giant has a long record of paying a large and growing dividend, and its yields have long topped the blue-chip average.

This in part reflects its exceptional cash generation, even during tough times. Today, its Solvency Capital Coverage Ratio is 172%, well inside its 140%-180% target. It may not protect Phoenix’s share price from falling if economic conditions worsen. But it at least means the company looks in good shape to pay this year’s expected dividends.

Over the longer term, I expect dividends here to rise steadily as demographic changes drive retirement product demand and push its profits skywards.

FTSE 250-listed SDCL Energy Efficiency Trust (LSE:SEIT) offers tantalising all-round value in my book. Its forward dividend yield is 10.8%, more than triple the Footsie average of 3.3%.

The investment trust also trades at a whopping 36.8% discount to its net asset value (NAV) per share of 91.5p.

Higher interest rates have weighed heavily on SDCL’s performance of late. With inflation edging upwards again, this is a threat that remains in play.

Yet I’m still confident in the trust’s long-term potential. It invests in projects that reduce heat wastage, improve on-site power generation, and cool commercial buildings, for example. And as such, it has considerable growth potential as companies try to meet their green targets.

On the dividend front, SDCL has raised shareholder payouts at an average rate of 4.8% over the last five years.

Launched in March 2024, the iShares World Equity High Income ETF (LSE:WINC) doesn’t have a long record of annual growth. But it’s tipped to raise the annual payout again this year, resulting in a vast 9.6% dividend yield.

The fund is designed “to generate income and capital growth with lower volatility than developed market equities“. To achieve a more stable performance than 100% share-based ETFs, some of its capital is also tied up in cash and government bonds.

That’s not to say that the fund’s fully protected from choppiness, however. Weighty exposure to cyclical sectors like information technology and financial services leaves it exposed to economic downturns.

However, it aims to counterbalance these with holdings of defensive shares like utilities, healthcare, telecoms, and consumer goods shares. As an investor, it’s also worth remembering iShares World Equity’s large contingent of high-growth shares (like Nvidia and Apple) also create potential for robust long-term dividend growth.

The post 7.9%+ dividend yields! Here are 3 high-yield income stocks, investment trusts, and ETFs to consider appeared first on The Motley Fool UK.

More reading

Royston Wild has positions in Legal & General Group Plc and Taylor Wimpey Plc. The Motley Fool UK has recommended Apple and Nvidia. Views expressed on the companies mentioned in this article are those of the writer and therefore may differ from the official recommendations we make in our subscription services such as Share Advisor, Hidden Winners and Pro. Here at The Motley Fool we believe that considering a diverse range of insights makes us better investors.

Motley Fool UK 2025