Microsatellite Instability (MSI-H) occurs in approximately 15% of non-metastatic colon cancers, influencing patient outcomes positively compared to microsatellite stable (MSS) cancers. This systematic review focuses on the prognostic significance of KRAS, NRAS, and BRAF mutations within MSI-H colon cancer. Through comprehensive searches in databases like MEDLINE, EMBASE, and others until 1 January 2024, we selected 8 pertinent studies from an initial pool of 1918. These studies, encompassing nine trials and five observational studies involving 13,273 patients, provided insights into disease-free survival (DFS), survival after recurrence, and overall survival. The pooled data suggest that while KRAS and BRAF mutations typically predict poorer outcomes in MSS colorectal cancer, their impact is less pronounced in MSI contexts, with implications varying across different stages of cancer and treatment responses. In particular, adverse effects of these mutations manifest significantly upon recurrence rather than affecting immediate DFS. Our findings confirm the complex interplay between genetic mutations and MSI status, emphasizing the nuanced role of MSI in modifying the prognostic implications of KRAS, NRAS, and BRAF mutations in colon cancer. This review underscores the importance of considering MSI alongside mutational status in the clinical decision-making process, aiming to tailor therapeutic strategies more effectively for colon cancer patients. Full article

Atypical atrial flutter (AFL) can be challenging to ablate, especially when involving dual-loop re-entry. We sought to assess the electroanatomical characteristics of single- and dual-loop AFLs in patients undergoing catheter ablation. Methods: We analyzed 25 non-cavotricuspid isthmus-dependent macro-re-entrant AFL in 19 consecutive patients. Three-dimensional high-density activation mapping was performed, and active re-entry loops were confirmed by entrainment mapping. Results: Of 25 AFLs (24 left, 1 right atrial), 13 (52%) exhibited dual-loop re-entry. The most common circuits included, in 6/13 (46% of dual loops), a perimitral re-entry with a second loop around the right/left pulmonary veins (PV) and, in 6/13 (46%), involved a right PV ostium with a second loop around either a functional conduction block or another PV. Ablation at the common isthmus of dual-loop AFLs and at the critical isthmus of single-loop AFLs terminated the arrhythmia more frequently than ablation at a secondary isthmus of dual-loop AFLs (5/6 (83%) and 8/11 (73%) versus 1/8 (13%), respectively, p = 0.013). Conclusion: More than half of AFLs exhibited a dual-loop re-entrant mechanism. Most critical isthmuses were found at the mitral isthmus, the left atrial roof or right PV ostia. Ablation targeting the common isthmus resulted in a higher termination rate. Full article

Non-small cell lung cancer (NSCLC) is the most common pulmonary malignancy, frequently diagnosed at an advanced stage (III/IV). Patients in the Locally Advanced Stage Subgroup (IIIA) are relatively few, yet compose heterogenic phenotypes, posing a diagnostic and treating challenge, leading to a lack of clinical guidelines regarding the optimal standard of care. Several approaches exist, with a general agreement that a combined oncological and surgical modality approach is required. In this current retrospective descriptive study, patients with operable stage IIIA NSCLC who underwent surgery between 2013 and 2020 were evaluated on several aspects, including the initial diagnosis, neoadjuvant regimens, outcomes of surgical intervention, and overall survival at 2 years and 5 years following treatment. A total of 35 patients had neoadjuvant oncological treatment (mostly chemoradiation therapy) prior to surgery, out of which 28 patients were diagnosed with stage IIIA NSCLC. In post-operative assessment of pathological staging, downstaging was reported in 19 patients, of which 25% of cases were defined as a complete pathological response. The 2-year overall survival rate was 65% and the 5-year overall survival rate was 62%. The main pattern of disease recurrence was distant metastasis. Full article

Focused parathyroidectomy is the preferred surgical method for treating primary hyperparathyroidism (pHPT) sustained by the pre-operatively well-localized parathyroid adenoma. We aimed to compare the effectiveness, safety, and short-term clinical outcome of minimally invasive video-assisted parathyroidectomy (MIVAP) in locoregional anaesthesia (LA) vs. general anaesthesia (GA) by means of propensity score matching (PSM) analysis. Retrospective research of patients who underwent MIVAP between January 2014 and December 2022 was carried out. Patients were divided into two groups based on the anaesthesiologic procedure (LA vs. GA). Overall, 553 patients underwent MIVAP. After PSM, 115 patients in the LA group and 230 patients in the GA group were included. MIVAP under LA was associated with shorter median operative time (16 vs. 35 min, p < 0.001), shorter median operative room occupation time (44 vs. 73 min, p < 0.001), and lesser median post-operative visual analogue scale pain, with comparable post-operative hospital stay and complication rate. MIVAP under LA is a safe and feasible procedure with significant advantages over GA in terms of post-operative pain and operative room occupation time. This last step can finally result in more efficient utilisation of the operative room and the health care system’s resources. Full article

With sustainability dominating the industry, recycling the generated waste from different processes is becoming increasingly important. This study focuses on recycling waste generated during aluminum anodizing waste (AAW) in friction material formulations for automotive braking applications. However, before utilization, the waste needs to be pre-treated, which mainly involves drying. Hence, four different industrial drying methods were studied to dry the AAW, and the corresponding characteristics were observed by evaluating its residual humidity and crushability index. The waste powders were further characterized using FT-IR and SEM/EDXS to understand their constituents. The initial analysis showed that the waste subjected to the drying process P2 and P1 with the lowest final humidity fetched the most desirable results, with P1 having the simpler drying procedure. The AAW powders were added in a commercial friction material formulation at 6 and 12 wt.% and subjected to friction, wear, and non-exhaust particulate matter analysis. The worn surfaces were analyzed using SEM/EDXS evaluation to understand the extension and composition of the deposited secondary contact plateaus. It was seen that the 12 wt.% addition of waste processed using the P1 technique provided the most satisfactory friction, wear, and emission characteristics, along with expansive secondary contact plateaus with a good contribution of the waste in its formation. This study showed a good relationship between the processing method and a formulation’s tribological and emission characteristics, thereby paving the way for using this drying method for other waste requiring pre-treatment. Full article

The optimal designing of the hybrid energy system (HES) is a challenging task due to the multiple objectives and various uncertainties. Especially for HES, primarily powered by solar energy, the reference solar radiation data directly impact the result of the optimization design. To incorporate the stochastic characteristics of solar radiation into the sizing process, a data-driven stochastic modeling method for solar radiation is proposed. The method involves two layers of stochastic processes that capture the intraday variation and daily evolution of solar radiation. First, the clearness index (CI) is introduced to describe the radiation intensity at different times. Then, the daily clearness state (DCS) is proposed, based on the statistical indicators of the intraday CI. The Markov model is used to describe the stochastic evolutionary characteristics between different DCSs. The probabilistic distribution of the CI under different DCS is obtained based on the diffusion kernel density estimation (DKDE), which is used for the stochastic generation of the CI at various times of the day. Finally, the radiation profile required for the optimal design is obtained by the stochastic generation of the DCS sequences and the intraday clearness index under corresponding states. A case study of an off-grid solar-powered HES is provided to illustrate this methodology. Full article

This review aims to analyze the technological and scientific applications regarding cocoa fermentation through a prospective study of patent documents and research articles. The Espacenet database was used as a patent research tool by searching both the IPC code “A23G1” and the terms “cocoa” and “ferment*”. A total of 130 documents were found—49 were related to the subject. The Scopus database was also searched for scientific articles using the terms “cocoa” and “fermentation”. A total of 812 articles were found—517 were related to the subject. Cocoa fermentation has not yet reached technological maturity, despite the growth in patent documents and scientific research observed in the last two decades. The creation of the Cacao of Excellence Program (2009), among others, has incentivized sustainability and quality in cocoa-producing countries. Brazil, Colombia, and Indonesia are leading with scientific publications in the last 5 years, despite the lack of patents filed. The United Kingdom, France, China, Canada, and Germany, despite not being cocoa-producing countries, are the main holders of the technology. Patent documents analyzed relate to food science, biotechnology, engineering, and chemistry. Microbial biotechnology has gained attention as a key factor to produce a higher-quality cocoa bean. Saccharomyces is the most frequent genus of yeast used as a starter culture in patent documents. Some patent documents propose the addition of fruits during cocoa fermentation, but a few scientific studies have been found on the matter. Overall, technological applications and scientific studies have focused on improving cocoa quality. The cocoa market is expected to increase significantly in the next few years, representing an opportunity to develop high-quality cocoa using novel fermentation techniques. Full article

This study measured parameters automatically by marking the point for measuring each parameter on whole-spine radiographs. Between January 2020 and December 2021, 1017 sequential lateral whole-spine radiographs were retrospectively obtained. Of these, 819 and 198 were used for training and testing the performance of the landmark detection model, respectively. To objectively evaluate the program’s performance, 690 whole-spine radiographs from four other institutions were used for external validation. The combined dataset comprised radiographs from 857 female and 850 male patients (average age 42.2 ± 27.3 years; range 20–85 years). The landmark localizer showed the highest accuracy in identifying cervical landmarks (median error 1.5–2.4 mm), followed by lumbosacral landmarks (median error 2.1–3.0 mm). However, thoracic landmarks displayed larger localization errors (median 2.4–4.3 mm), indicating slightly reduced precision compared with the cervical and lumbosacral regions. The agreement between the deep learning model and two experts was good to excellent, with intraclass correlation coefficient values >0.88. The deep learning model also performed well on the external validation set. There were no statistical differences between datasets in all parameters, suggesting that the performance of the artificial intelligence model created was excellent. The proposed automatic alignment analysis system identified anatomical landmarks and positions of the spine with high precision and generated various radiograph imaging parameters that had a good correlation with manual measurements. Full article

Underwater images suffer from low contrast and color distortion. In order to improve the quality of underwater images and reduce storage and computational resources, this paper proposes a lightweight model Rep-UWnet to enhance underwater images. The model consists of a fully connected convolutional network and three densely connected RepConv blocks in series, with the input images connected to the output of each block with a Skip connection. First, the original underwater image is subjected to feature extraction by the SimSPPF module and is processed through feature summation with the original one to be produced as the input image. Then, the first convolutional layer with a kernel size of 3 × 3, generates 64 feature maps, and the multi-scale hybrid convolutional attention module enhances the useful features by reweighting the features of different channels. Second, three RepConv blocks are connected to reduce the number of parameters in extracting features and increase the test speed. Finally, a convolutional layer with 3 kernels generates enhanced underwater images. Our method reduces the number of parameters from 2.7 M to 0.45 M (around 83% reduction) but outperforms state-of-the-art algorithms by extensive experiments. Furthermore, we demonstrate our Rep-UWnet effectively improves high-level vision tasks like edge detection and single image depth estimation. This method not only surpasses the contrast method in objective quality, but also significantly improves the contrast, colorimetry, and clarity of underwater images in subjective quality. Full article

Quantum tunneling in a two-dimensional integrable map is studied. The orbits of the map are all confined to the curves specified by the one-dimensional Hamiltonian. It is found that the behavior of tunneling splitting for the integrable map and the associated Hamiltonian system is qualitatively the same, with only a slight difference in magnitude. However, the tunneling tails of the wave functions, obtained by superposing the eigenfunctions that form the doublet, exhibit significant differences. To explore the origin of the difference, we observe the classical dynamics in the complex plane and find that the existence of branch points appearing in the potential function of the integrable map could play the role of yielding non-trivial behavior in the tunneling tail. The result highlights the subtlety of quantum tunneling, which cannot be captured in nature only by the dynamics in the real plane. Full article

Clarifying the relationship between urbanization and farmland abandonment in urban agglomerations (UAs) is crucial to guide the formulation of arable land management policies and strategies for sustainable urban development. Despite numerous studies confirming the correlation between farmland abandonment and certain urbanization factors, the exploration of the patterns and underlying mechanisms of farmland abandonment in China’s UAs remains worthy of systematic investigation. In this study, we conducted an analysis of the spatiotemporal trends in farmland abandonment and examined the key drivers of farmland abandonment in four representative Chinese UAs—Beijing–Tianjin–Hebei (BTH), Chengdu–Chongqing (CC), Pearl River Delta (PRD), and Yangtze River Delta (YRD). Our findings reveal that farmland abandonment has been intensified with increasing fragmentation and aggregation patches across these UAs. Abandonment experience was the main driver of continuous abandonment. Moreover, natural conditions persistently influenced farmland abandonment in the BTH, while land urbanization and economic urbanization were predominant drivers in the CC. The abandonment in the PRD was mainly driven by population urbanization, while the abandonment in the YRD was primarily driven by economic urbanization and land urbanization. The research findings provide data support and scientific explanation for land policy-making in these typical UAs under different development strategies. Full article

The Ca Mau Peninsula, situated in the Mekong Delta of Vietnam, features low-lying terrain. In addition to the challenges posed by climate change, land subsidence in the area is exacerbated by the overexploitation of groundwater and intensive agricultural practices. In this study, we assessed the land subsidence susceptibility in the Ca Mau Peninsula utilizing three boosting machine learning models: AdaBoost, Gradient Boosting, and Extreme Gradient Boosting (XGB). Eight key factors were identified as the most influential in land subsidence within Ca Mau: land cover (LULC), groundwater depth, digital terrain model (DTM), normalized vegetation index (NDVI), geology, soil composition, distance to roads, and distance to rivers and streams. The dataset includes 2011 points referenced from the Persistent Scattering SAR Interferometry (PSI) method, of which 1011 points are subsidence points and the remaining are non-subsidence points. The sample points were split, with 70% allocated to the training set and 30% to the testing set. Following computation and execution, the three models underwent evaluation for accuracy using statistical metrics such as the receiver operating characteristic (ROC) curve, area under the curve (AUC), specificity, sensitivity, and overall accuracy (ACC). The research findings revealed that the XGB model exhibited the highest accuracy, achieving an AUC and ACC above 0.88 for both the training and test sets. Consequently, XGB was chosen to construct a land subsidence susceptibility map for the Ca Mau Peninsula. In addition, 31 subsidence points measured by leveling surveys between 2005 and 2020, provided by the Department of Survey, Mapping and Geographic Information Vietnam, were used for validating the land subsidence susceptibility from the XGB method. The findings indicate a 70.9% accuracy rate in predicting subsidence susceptibility compared to the leveling measurement points. Full article

Background: Leukoencephalopathy with brainstem and spinal cord involvement and lactate elevation is an inherited disease caused by pathogenic biallelic variants in the gene DARS2, which encodes mitochondrial aspartyl-tRNA synthetase. This disease is characterized by slowly progressive spastic gait, cerebellar symptoms, and leukoencephalopathy with brainstem and spinal cord involvement. Case Presentation: Peripheral blood samples were collected from four patients from four unrelated families to extract genomic DNA. All patients underwent partial exon analysis of the DARS2 gene using Sanger sequencing, which detected the c.228-21_228-20delinsC variant in a heterozygous state. Further DNA from three patients was analyzed using a next-generation sequencing-based custom AmpliSeq™ panel for 59 genes associated with leukodystrophies, and one of the patients underwent whole genome sequencing. We identified a novel pathogenic variant c.1675-1256_*115delinsGCAACATTTCGGCAACATTCCAACC in the DARS2 gene. Three patients (patients 1, 2, and 4) had slowly progressive cerebellar ataxia, and two patients (patients 1 and 2) had spasticity. In addition, two patients (patients 2 and 4) showed signs of axonal neuropathy, such as decreased tendon reflexes and loss of distal sensitivity. Three patients (patients 1, 2, and 3) also had learning difficulties. It should be noted the persistent presence of characteristic changes in brain MRI in all patients, which emphasizes its importance as the main diagnostic tool for suspicion and subsequent confirmation of LBSL. Conclusions: We found a novel indel variant in the DARS2 gene in four patients with LBSL and described their clinical and genetic characteristics. These results expand the mutational spectrum of LBSL and aim to improve the laboratory diagnosis of this form of leukodystrophy. Full article

With the surging interest in electric vehicles (EVs), there is a need for advancements in the development and dismantling of lithium-ion batteries (LIBs), which are highly important for the circular economy. This paper introduces an intelligent hybrid task planner designed for multi-robot disassembly and demonstrates its application to an EV lithium-ion battery pack. The objective is to enable multiple robots to operate collaboratively in a single workspace to execute battery disassembly tasks efficiently and without collisions. This approach can be generalized to almost any disassembly task. The planner uses logical and hierarchical strategies to identify object locations from data captured by cameras mounted on each robot’s end-effector, orchestrating coordinated pick-and-place operations. The efficacy of this task planner was assessed through simulations with three trajectory-planning algorithms: RRT, RRTConnect, and RRTStar. Performance evaluations focused on completion times for battery disassembly tasks. The results showed that completion times were similar across the planners, with 543.06 s for RRT, 541.89 s for RRTConnect, and 547.27 s for RRTStar, illustrating that the effectiveness of the task planner is independent of the specific joint-trajectory-planning algorithm used. This demonstrates the planner’s capability to effectively manage multi-robot disassembly operations. Full article

Highly interactive honeypots can form reliable connections by responding to attackers to delay and capture intranet attacks. However, current research focuses on modeling the attacker as part of the environment and defining single-step attack actions by simulation to study the interaction of honeypots. It ignores the iterative nature of the attack and defense game, which is inconsistent with the correlative and sequential nature of actions in real attacks. These limitations lead to insufficient interaction of the honeypot response strategies generated by the study, making it difficult to support effective and continuous games with attack behaviors. In this paper, we propose an autonomous attack response framework (named AARF) to enhance interaction based on multi-agent dynamic games. AARF consists of three parts: a virtual honeynet environment, attack agents, and defense agents. Attack agents are modeled to generate multi-step attack chains based on a Hidden Markov Model (HMM) combined with the generic threat framework ATT&CK (Adversarial Tactics, Techniques, and Common Knowledge). The defense agents iteratively interact with the attack behavior chain based on reinforcement learning (RL) to learn to generate honeypot optimal response strategies. Aiming at the sample utilization inefficiency problem of random uniform sampling widely used in RL, we propose the dynamic value label sampling (DVLS) method in the dynamic environment. DVLS can effectively improve the sample utilization during the experience replay phase and thus improve the learning efficiency of honeypot agents under the RL framework. We further couple it with a classic DQN to replace the traditional random uniform sampling method. Based on AARF, we instantiate different functional honeypot models for deception in intranet scenarios. In the simulation environment, honeypots collaboratively respond to multi-step intranet attack chains to defend against these attacks, which demonstrates the effectiveness of AARF. The average cumulative reward of the DQN with DVLS is beyond eight percent, and the convergence speed is improved by five percent compared to a classic DQN. Full article

Previous studies have shown encouraging results regarding the efficacy and safety of nutraceuticals, such as “red yeast rice (RYR) extract”, on reducing hypercholesterolemia in humans. A systematic review and meta-analysis was conducted from January 2012 to May 2022. The search was strictly focused on clinical trials that examined the association between RYR extract consumption and parameters of the lipid profile in humans. Fourteen double-blinded clinical trials were identified. The interventions lasted 4–24 weeks. In most studies, there was one intervention group and one control group. RYR extract consumption statistically significantly reduced total cholesterol (mean absolute reduction: 37.43 mg/dL; 95% confidence interval [CI]: −47.08, −27.79) and low-density lipoprotein cholesterol (LDL-C; mean absolute reduction: 35.82 mg/dL; 95% CI: −43.36, −28.29), but not high-density lipoprotein cholesterol, triglycerides and apolipoproteins A-I and B. As regards the safety, RYR extract was considered a safe choice with neither threatening nor frequent side effects. The consumption of RYR extract by people with hypercholesterolemia was associated with statistically significant reduction in total cholesterol and LDL-C, whereas it was not associated with an increase in life-threatening side effects. Further research on specific subpopulations and outcomes could establish a consensus on determining the clinical benefits and potential risks, if any, of this nutraceutical. Full article

Due to their advantages of good stability, adaptability, and flexibility, multi-legged robots are increasingly important in fields such as rescue, military, and healthcare. This study focuses on the millipede, a multi-segmented organism, and designs a novel multi-segment biomimetic robot based on an in-depth investigation of the millipede’s biological characteristics and locomotion mechanisms. Key leg joints of millipede locomotion are targeted, and a mathematical model of the biomimetic robot’s leg joint structure is established for kinematic analysis. Furthermore, a central pattern generator (CPG) control strategy is studied for multi-jointed biomimetic millipede robots. Inspired by the millipede’s neural system, a simplified single-loop CPG network model is constructed, reducing the number of oscillators from 48 to 16. Experimental trials are conducted using a prototype to test walking in a wave-like gait, walking with a leg removed, and walking on complex terrain. The results demonstrate that under CPG waveform input conditions, the robot can walk stably, and the impact of a leg failure on overall locomotion is acceptable, with minimal speed loss observed when walking on complex terrain. The research on the structure and motion control algorithms of multi-jointed biomimetic robots lays a technical foundation, expanding their potential applications in exploring unknown environments, rescue missions, agriculture, and other fields. Full article

Endometrial cancer (EC) poses a significant health issue among women, and its incidence has been rising for a couple of decades. Surgery remains its principal treatment method and may have a curative, staging, or palliative aim. The type and extent of surgery depends on many factors, and the risks and benefits should be carefully weighed. While simple hysterectomy might be sufficient in early stage EC, modified-radical hysterectomy is sometimes indicated. In advanced disease, the evidence suggests that, similarly to ovarian cancer, optimal cytoreduction improves survival rate. The role of lymphadenectomy in EC patients has long been a controversial issue. The rationale for systematic lymphadenectomy and the procedure of the sentinel lymph node biopsy are thoroughly discussed. Finally, the impact of the molecular classification and new International Federation of Gynecology and Obstetrics (FIGO) staging system on EC treatment is outlined. Due to the increasing knowledge on the pathology and molecular features of EC, as well as the new advances in the adjuvant therapies, the surgical management of EC has become more complex. In the modern approach, it is essential to adjust the extent of the surgery to a specific patient, ensuring an optimal, made-to-measure personalized surgery. This narrative review focuses on the intricacies of surgical management of EC and aims at summarizing the available literature on the subject, providing an up-to-date clinical guide. Full article

The extracellular matrix (ECM) is composed of complex fibrillar proteins, proteoglycans, and macromolecules, generated by stromal, immune, and cancer cells. The components and organisation of the matrix evolves as tumours progress to invasive disease and metastasis. In many solid tumours, dense fibrotic ECM has been hypothesised to impede therapy response by limiting drug and immune cell access. Interventions to target individual components of the ECM, collectively termed the matrisome, have, however, revealed complex tumour-suppressor, tumour-promoter, and immune-modulatory functions, which have complicated clinical translation. The degree to which distinct components of the matrisome can dictate tumour phenotypes and response to therapy is the subject of intense study. A primary aim is to identify therapeutic opportunities within the matrisome, which might support a better response to existing therapies. Many matrix signatures have been developed which can predict prognosis, immune cell content, and immunotherapy responses. In this review, we will examine key components of the matrisome which have been associated with advanced tumours and therapy resistance. We have primarily focussed here on targeting matrisome components, rather than specific cell types, although several examples are described where cells of origin can dramatically affect tumour roles for matrix components. As we unravel the complex biochemical, biophysical, and intracellular transduction mechanisms associated with the ECM, numerous therapeutic opportunities will be identified to modify tumour progression and therapy response. Full article

Background: Clinical case illustrations of patients with an impairment of personality functioning (IPF) have repeatedly reported that progress during psychotherapy is reflected by alterations in dream content. However, quantitative studies based on samples of psychotherapy patients are scarce. As a core component of both personality functioning and contemporary psychodynamic dream theory, the construct of affect regulation is of specific significance in this context. Aims: To test if improvement in personality functioning in the course of psychotherapy is associated with an increasing ability to regulate affects in dreams. Method: In a longitudinal design, affect regulation was compared in N = 94 unsolicited dream reports from the first vs. last third of long term psychotherapy of ten patients with initial IPF. Dream reports were transcribed from recordings of the sessions. Expert ratings of the level of personality functioning were obtained using the Scales of Psychological Capacities. The capacity for affect regulation was assessed using the Zurich Dream Process Coding System. Group differences were assessed using linear mixed models, controlling for dream length as well as the nested structure of this data set. Results: Patients demonstrated an increased capacity for affect regulation in dreams that was primarily evident in three core features: the complexity of dream elements (cf., e.g., parameter attributes, p = 0.024); the extent of affective involvement in the dream ego (cf., e.g., parameter subject feeling, p = 0.014); and the flexibility to regulate the dynamics of safety/involvement processes (p = < 0.001). This pattern was especially prominent in a subgroup (n = 7) of patients with more pronounced improvements in personality functioning. Conclusion: These findings support the hypotheses that decreasing IPF during psychotherapy is associated with increases in the capacity for affect regulation in dreams. Thus, researchers and therapists can utilize dream reports to illuminate the important aspects of treatment progress in clinical practice. Full article

To improve the power generation efficiency of offshore wind turbines and address the problem of high fire monitoring and warning costs, we propose a data-driven fire warning method based on transfer learning for wind turbines in this paper. This paper processes wind turbine operation data in a SCADA system. It uses an extreme gradient-boosting tree (XGBoost) algorithm to build an offshore wind turbine unit fire warning model with a multiparameter prediction function. This paper selects some parameters from the dataset as input variables for the model, with average cabin temperature, average outdoor temperature, average cabin humidity, and average atmospheric humidity as output variables. This paper analyzes the distribution information of input and output variables and their correlation, analyzes the predicted difference, and then provides an early warning for wind turbine fires. This paper uses this fire warning model to transfer learning to different models of offshore wind turbines in the same wind farm to achieve fire warning. The experimental results show that the prediction performance of the multiparameter is accurate, with an average MAPE of 0.016 and an average RMSE of 0.795. It is better than the average MAPE (0.051) and the average RMSE (2.020) of the prediction performance of a backpropagation (BP) neural network, as well as the average MAPE (0.030) and the average RMSE (1.301) of the prediction performance of random forest. The transfer learning model has good prediction performance, with an average MAPE of 0.022 and an average RMSE of 1.469. Full article

Federated learning (FL) has emerged as one of the de-facto privacy-preserving paradigms that can effectively work with decentralized data sources (e.g., hospitals) without acquiring any private data. Recently, applications of FL have vastly expanded into multiple domains, particularly the medical domain, and FL is becoming one of the mainstream technologies of the near future. In this study, we provide insights into FL fundamental concepts (e.g., the difference from centralized learning, functions of clients and servers, workflows, and nature of data), architecture and applications in the general medical domain, synergies with emerging technologies, key challenges (medical domain), and potential research prospects. We discuss major taxonomies of the FL systems and enlist technical factors in the FL ecosystem that are the foundation of many adversarial attacks on these systems. We also highlight the promising applications of FL in the medical domain by taking the recent COVID-19 pandemic as an application use case. We highlight potential research and development trajectories to further enhance the persuasiveness of this emerging paradigm from the technical point of view. We aim to concisely present the progress of FL up to the present in the medical domain including COVID-19 and to suggest future research trajectories in this area. Full article