QSAR: Using the Past to Study the Present.

Quantitative structure-activity relationships (QSAR) is a method for predicting the physical and biological properties of small molecules; it is in use in industry and public services. However, as any scientific method, it is challenged by more and more requests, especially considering its possible role in assessing the safety of new chemicals. To answer the question whether QSAR, by exploiting available knowledge, can build new knowledge, the chapter reviews QSAR methods in search of a QSAR epistemology. QSAR stands on tree pillars, i.e., biological data, chemical knowledge, and modeling algorithms. Usually the biological data, resulting from good experimental practice, are taken as a true picture of the world; chemical knowledge has scientific bases; so if a QSAR model is not working, blame modeling. The role of modeling in developing scientific theories, and in producing knowledge, is so analyzed. QSAR is a mature technology and is part of a large body of in silico methods and other computational methods. The active debate about the acceptability of the QSAR models, about the way to communicate them, and the explanation to provide accompanies the development of today QSAR models. An example about predicting possible endocrine-disrupting chemicals (EDC) shows the many faces of modern QSAR methods.

Machine learning in oncological pharmacogenomics: advancing personalized chemotherapy.

This review analyzes the application of machine learning (ML) in oncological pharmacogenomics, focusing on customizing chemotherapy treatments. It explores how ML can analyze extensive genomic, proteomic, and other omics datasets to identify genetic patterns associated with drug responses. This, in turn, facilitates personalized therapies that are more effective and have fewer side effects. Recent studies have emphasized ML's revolutionary role of ML in personalized oncology treatment by identifying genetic variability and understanding cancer pharmacodynamics. Integrating ML with electronic health records and clinical data shows promise in refining chemotherapy recommendations by considering the complex influencing factors. Although standard chemotherapy depends on population-based doses and treatment regimens, customized techniques use genetic information to tailor treatments for specific patients, potentially enhancing efficacy and reducing adverse effects.However, challenges, such as model interpretability, data quality, transparency, ethical issues related to data privacy, and health disparities, remain. Machine learning has been used to transform oncological pharmacogenomics by enabling personalized chemotherapy treatments. This review highlights ML's potential of ML to enhance treatment effectiveness and minimize side effects through detailed genetic analysis. It also addresses ongoing challenges including improved model interpretability, data quality, and ethical considerations. The review concludes by emphasizing the importance of rigorous clinical trials and interdisciplinary collaboration in the ethical implementation of ML-driven personalized medicine, paving the way for improved outcomes in cancer patients and marking a new frontier in cancer treatment.

Modeling of the effect of gradation and compaction characteristics on the california bearing ratio of granular materials for subbase and landfill liner construction.

The California bearing ratio (CBR) of a granular materials are influence by the soil particle distribution indices such as D10, D30, D50, and D60 and also the compaction properties such as the maximum dry density (MDD) and the optimum moisture content (OMC). For this reason, the particle packing and compactibility of the soil play a big role in the design and construction of subbases and landfills. In this research paper, experimental data entries have been collected reflecting the CBR behavior of granular soil used to construct landfill and subbase. The database was utilized in the ratio of 78-22% to predict the CBR behavior considering the artificial neural network (ANN), the evolutionary polynomial regression (EPR), the genetic programming (GP), Extreme Gradient Boosting (XGBoost), Random Forest (RF) and the response surface methodology (RSM) intelligent learning and symbolic abilities. The relative importance values for each input parameter were carried out, which indicated that the (CBR) value depends mainly on the average particle size (D30, 50 & 60). They showed a combined influence index of 66% of the considered parameters in the model exercise. This further shows the importance and structural influence of the particles within the D50 and D60 range in a granular material consistency in the design and construction purposes. Performance indices were also used to study the ability of the models. The ANN model showed the best performance with accuracy of 88%, then GP, EPR and RF with almost the same accuracies of 85% and lastly the XGBoost with accuracy of 81%. Also, the RSM produced an R2 of 0.9464 with a p-value of less than 0.0001. These values show that the ANN produced the decisive model with the superior performance indices in the forecast of CBR of granular material used as subbase and waste compacted earth liner material. The results further show that optimal performance of the CBR depended on D50 and D60 for the design of subgrade, subbase, and liner purposes and also during the performance monitoring phase of the constructed flexible pavement foundations and compacted earth liners.

Development and application of explainable artificial intelligence using machine learning classification for long-term facial nerve function after vestibular schwannoma surgery.

PURPOSE: Vestibular schwannomas (VSs) represent the most common cerebellopontine angle tumors, posing a challenge in preserving facial nerve (FN) function during surgery. We employed the Extreme Gradient Boosting machine learning classifier to predict long-term FN outcomes (classified as House-Brackmann grades 1-2 for good outcomes and 3-6 for bad outcomes) after VS surgery. METHODS: In a retrospective analysis of 256 patients, comprehensive pre-, intra-, and post-operative factors were examined. We applied the machine learning (ML) classifier Extreme Gradient Boosting (XGBoost) for the following binary classification: long-term good and bad FN outcome after VS surgery To enhance the interpretability of our model, we utilized an explainable artificial intelligence approach. RESULTS: Short-term FN function (tau = 0.6) correlated with long-term FN function. The model exhibited an average accuracy of 0.83, a ROC AUC score of 0.91, and Matthew's correlation coefficient score of 0.62. The most influential feature, identified through SHapley Additive exPlanations (SHAP), was short-term FN function. Conversely, large tumor volume and absence of preoperative auditory brainstem responses were associated with unfavorable outcomes. CONCLUSIONS: We introduce an effective ML model for classifying long-term FN outcomes following VS surgery. Short-term FN function was identified as the key predictor of long-term function. This model's excellent ability to differentiate bad and good outcomes makes it useful for evaluating patients and providing recommendations regarding FN dysfunction management.

"Enhancing post-craniotomy recovery: leveraging AI and network analysis for improved outcomes".

This letter addresses the importance of enhancing post-craniotomy care for primary brain tumor patients by leveraging insights from Rongqing Li et al.'s study on symptom networks. The study identified key central and bridge symptoms, such as sadness and difficulty understanding, which influence post-surgical recovery and quality of life. It also highlighted that patients with noninvasive tumors showed more cohesive symptom networks compared to those with invasive tumors. However, the study had limitations, including a short observation period and reliance on self-reported data, which restricted the depth of the findings.To optimize recovery, integrating artificial intelligence (AI) and machine learning (ML) could revolutionize post-craniotomy care. AI can assist with surgical planning, predict complications, and monitor recovery through wearable devices and real-time alerts. Natural Language Processing (NLP) can improve symptom detection from electronic health records, enhancing clinical decision-making. Despite the potential of these technologies, ethical concerns regarding data privacy and AI-generated report accuracy must be addressed. Future research should focus on long-term outcomes and refining AI applications to improve post-craniotomy symptom management and overall patient outcomes.

Exploring metabolic anomalies in COVID-19 and post-COVID-19: a machine learning approach with explainable artificial intelligence.

The COVID-19 pandemic, caused by SARS-CoV-2, has led to significant challenges worldwide, including diverse clinical outcomes and prolonged post-recovery symptoms known as Long COVID or Post-COVID-19 syndrome. Emerging evidence suggests a crucial role of metabolic reprogramming in the infection's long-term consequences. This study employs a novel approach utilizing machine learning (ML) and explainable artificial intelligence (XAI) to analyze metabolic alterations in COVID-19 and Post-COVID-19 patients. Samples were taken from a cohort of 142 COVID-19, 48 Post-COVID-19, and 38 control patients, comprising 111 identified metabolites. Traditional analysis methods, like PCA and PLS-DA, were compared with ML techniques, particularly eXtreme Gradient Boosting (XGBoost) enhanced by SHAP (SHapley Additive exPlanations) values for explainability. XGBoost, combined with SHAP, outperformed traditional methods, demonstrating superior predictive performance and providing new insights into the metabolic basis of the disease's progression and aftermath. The analysis revealed metabolomic subgroups within the COVID-19 and Post-COVID-19 conditions, suggesting heterogeneous metabolic responses to the infection and its long-term impacts. Key metabolic signatures in Post-COVID-19 include taurine, glutamine, alpha-Ketoglutaric acid, and LysoPC a C16:0. This study highlights the potential of integrating ML and XAI for a fine-grained description in metabolomics research, offering a more detailed understanding of metabolic anomalies in COVID-19 and Post-COVID-19 conditions.

Can We Simplify Liposome Manufacturing Using a Complex DoE Approach?

Microfluidic liposome production presents a streamlined pathway for expediting the translation of liposomal formulations from the laboratory setting to clinical applications. Using this production method, resultant liposome characteristics can be tuned through the control of both the formulation parameters (including the lipids and solvents used) and production parameters (including the production speed and mixing ratio). Therefore, the aim of this study was to investigate the relationship between not only total flow rate (TFR), the fraction of the aqueous flow rate over the organic flow rate (flow rate ratio (FRR)), and the lipid concentration, but also the solvent selection, aqueous buffer, and production temperature. To achieve this, we used temperature, applying a design of experiment (DoE) combined with machine learning. This study demonstrated that liposome size and polydispersity were influenced by manipulation of not only the total flow rate and flow rate ratio but also through the lipids, lipid concentration, and solvent selection, such that liposome attributes can be in-process controlled, and all factors should be considered within a manufacturing process as impacting on liposome critical quality attributes.

Machine learning prediction of no reflow in patients with ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention.

Background: No-reflow (NRF) phenomenon is a significant challenge in patients with ST-segment elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention (pPCI). Accurate prediction of NRF may help improve clinical outcomes of patients. This retrospective study aimed at creating an optimal model based on machine learning (ML) to predict NRF in these patients, with the additional objective of guiding pre- and intra-operative decision-making to reduce NRF incidence. Methods: Data were collected from 321 STEMI patients undergoing pPCI between January 2022 and May 2023, with the dataset being randomly divided into training and internal validation sets in a 7:3 ratio. Selected features included pre- and intra-operative demographic data, laboratory parameters, electrocardiogram, comorbidities, patients' clinical status, coronary angiographic data, and intraoperative interventions. Post comprehensive feature cleaning and engineering, three logistic regression (LR) models [LR-classic, LR-random forest (LR-RF), and LR-eXtreme Gradient Boosting (LR-XGB)], a RF model and an eXtreme Gradient Boosting (XGBoost) model were developed within the training set, followed by performance evaluation on the internal validation sets. Results: Among the 261 patients who met the inclusion criteria, 212 were allocated to the normal flow group and 49 to the NRF group. The training group consisted of 183 patients, while the internal validation group included 78 patients. The LR-XGB model, with an area under the curve (AUC) of 0.829 [95% confidence interval (CI): 0.779-0.880], was selected as the representative model for logistic regression analyses. The LR model had an AUC slightly lower than XGBoost model (AUC 0.835, 95% CI: 0.781-0.889) but significantly higher than RF model (AUC 0.731, 95% CI: 0.660-0.802). Internal validation underscored the unique advantages of each model, with the LR model demonstrating the highest clinical net benefit at relevant thresholds, as determined by decision curve analysis. The LR model encompassed seven meaningful features, and notably, thrombolysis in myocardial infarction flow after initial balloon dilation (TFAID) was the most impactful predictor in all models. A web-based application based on the LR model, hosting these predictive models, is available at https://l7173o-wang-lyn.shinyapps.io/shiny-1/. Conclusions: A LR model was successfully developed through ML to forecast NRF phenomena in STEMI patients undergoing pPCI. A web-based application derived from the LR model facilitates clinical implementation.

Drone and ground-truth data collection, image annotation and machine learning: A protocol for coastal habitat mapping and classification.

Aerial drone imaging is an efficient tool for mapping and monitoring of coastal habitats at high spatial and temporal resolution. Specifically, drone imaging allows for time- and cost-efficient mapping covering larger areas than traditional mapping and monitoring techniques, while also providing more detailed information than those from airplanes and satellites, enabling for example to differentiate various types of coastal vegetation. Here, we present a systematic method for shallow water habitat classification based on drone imagery. The method includes:•Collection of drone images and creation of orthomosaics.•Gathering ground-truth data in the field to guide the image annotation and to validate the final map product.•Annotation of drone images into - potentially hierarchical - habitat classes and training of machine learning algorithms for habitat classification.As a case study, we present a field campaign that employed these methods to map a coastal site dominated by seagrass, seaweed and kelp, in addition to sediments and rock. Such detailed but efficient mapping and classification can aid to understand and sustainably manage ecologically and valuable marine ecosystems.

Machine learning identifies a 5-serum cytokine panel for the early detection of chronic atrophy gastritis patients.

BACKGROUND: Chronic atrophy gastritis (CAG) is a high-risk pre-cancerous lesion for gastric cancer (GC). The early and accurate detection and discrimination of CAG from benign forms of gastritis (e.g. chronic superficial gastritis, CSG) is critical for optimal management of GC. However, accurate non-invasive methods for the diagnosis of CAG are currently lacking. Cytokines cause inflammation and drive cancer transformation in GC, but their utility as a diagnostic for CAG is poorly characterized. METHODS: Blood samples were collected, and 40 cytokines were quantified using a multiplexed immunoassay from 247 patients undergoing screening via endoscopy. Patients were divided into discovery and validation sets. Each cytokine importance was ranked using the feature selection algorithm Boruta. The cytokines with the highest feature importance were selected for machine learning (ML), using the LightGBM algorithm. RESULTS: Five serum cytokines (IL-10, TNF-α, Eotaxin, IP-10 and SDF-1a) that could discriminate between CAG and CSG patients were identified and used for predictive model construction. This model was robust and could identify CAG patients with high performance (AUC = 0.88, Accuracy = 0.78). This compared favorably to the conventional approach using the PGI/PGII ratio (AUC = 0.59). CONCLUSION: Using state-of-the-art ML and a blood-based immunoassay, we developed an improved non-invasive screening method for the detection of precancerous GC lesions. FUNDING: Supported in part by grants from: Jiangsu Science and Technology Project (no. BK20211039); Top Talent Support Program for young and middle-aged people of Wuxi Health Committee (BJ2023008); Medical Key Discipline Program of Wuxi Health Commission (ZDXK2021010), Wuxi Science and Technology Bureau Project (no. N20201004); Scientific Research Program of Wuxi Health Commission (Z202208, J202104).

A Chronicle Review of In-Silico Approaches for Discovering Novel Antimicrobial Agents to Combat Antimicrobial Resistance.

Antimicrobial resistance (AMR) poses a foremost threat to global health, necessitating innovative strategies for discovering antimicrobial agents. This review explores the role and recent advances of in-silico techniques in identifying novel antimicrobial agents and combating AMR giving few briefings of recent case studies of AMR. In-silico techniques, such as homology modeling, virtual screening, molecular docking, pharmacophore modeling, molecular dynamics simulation, density functional theory, integrated machine learning, and artificial intelligence, are systematically reviewed for their utility in discovering antimicrobial agents. These computational methods enable the rapid screening of large compound libraries, prediction of drug-target interactions, and optimization of drug candidates. The review discusses integrating in-silico approaches with traditional experimental methods and highlights their potential to accelerate the discovery of new antimicrobial agents. Furthermore, it emphasizes the significance of interdisciplinary collaboration and data-sharing initiatives in advancing antimicrobial research. Through a comprehensive discussion of the latest developments in in-silico techniques, this review provides valuable insights into the future of antimicrobial research and the fight against AMR. Supplementary Information: The online version contains supplementary material available at 10.1007/s12088-024-01355-x.

Unlocking Hidden Risks: Harnessing Artificial Intelligence (AI) to Detect Subclinical Conditions from an Electrocardiogram (ECG).

Recent artificial intelligence (AI) advancements in cardiovascular medicine offer potential enhancements in diagnosis, prediction, treatment, and outcomes. This article aims to provide a basic understanding of AI enabled ECG technology. Specific conditions and findings will be discussed, followed by reviewing associated terminology and methodology. In the appendix, definitions of AUC versus accuracy are explained. The application of deep learning models enables detecting diseases from normal electrocardiograms at accuracy not previously achieved by technology or human experts. Results with AI enabled ECG are encouraging as they considerably exceeded current screening models for specific conditions (i.e., atrial fibrillation, left ventricular dysfunction, aortic stenosis, and hypertrophic cardiomyopathy). This could potentially lead to a revitalization of the utilization of the ECG in the insurance domain. While we are embracing the findings with this rapidly evolving technology, but cautious optimism is still necessary at this point.

AI-Enhanced ECG Applications in Cardiology: Comprehensive Insights from the Current Literature with a Focus on COVID-19 and Multiple Cardiovascular Conditions.

The application of artificial intelligence (AI) in electrocardiography is revolutionizing cardiology and providing essential insights into the consequences of the COVID-19 pandemic. This comprehensive review explores AI-enhanced ECG (AI-ECG) applications in risk prediction and diagnosis of heart diseases, with a dedicated chapter on COVID-19-related complications. Introductory concepts on AI and machine learning (ML) are explained to provide a foundational understanding for those seeking knowledge, supported by examples from the literature and current practices. We analyze AI and ML methods for arrhythmias, heart failure, pulmonary hypertension, mortality prediction, cardiomyopathy, mitral regurgitation, hypertension, pulmonary embolism, and myocardial infarction, comparing their effectiveness from both medical and AI perspectives. Special emphasis is placed on AI applications in COVID-19 and cardiology, including detailed comparisons of different methods, identifying the most suitable AI approaches for specific medical applications and analyzing their strengths, weaknesses, accuracy, clinical relevance, and key findings. Additionally, we explore AI's role in the emerging field of cardio-oncology, particularly in managing chemotherapy-induced cardiotoxicity and detecting cardiac masses. This comprehensive review serves as both an insightful guide and a call to action for further research and collaboration in the integration of AI in cardiology, aiming to enhance precision medicine and optimize clinical decision-making.

The Global Evolution and Impact of Systems Biology and Artificial Intelligence in Stem Cell Research and Therapeutics Development: A Scoping Review.

Advanced bioinformatics analysis, such as systems biology (SysBio) and artificial intelligence (AI) approaches, including machine learning (ML) and deep learning (DL), is increasingly present in stem cell (SC) research. An approximate timeline on these developments and their global impact is still lacking. We conducted a scoping review on the contribution of SysBio and AI analysis to SC research and therapy development based on literature published in PubMed between 2000 and 2024. We identified an 8-10-fold increase in research output related to all three search terms between 2000 and 2021, with a 10-fold increase in AI-related production since 2010. Use of SysBio and AI still predominates in preclinical basic research with increasing use in clinically oriented translational medicine since 2010. SysBio- and AI-related research was found all over the globe, with SysBio output led by the United States (US, n=1487), United Kingdom (UK, n=1094), Germany (n=355), The Netherlands (n=339), Russia (n=215), and France (n=149), while for AI-related research the US (n=853) and UK (n=258) take a strong lead, followed by Switzerland (n=69), The Netherlands (n=37), and Germany (n=19). The US and UK are most active in SCs publications related to AI/ML and AI/DL. The prominent use of SysBio in ESC research was recently overtaken by prominent use of AI in iPSC and MSC research. This study reveals the global evolution and growing intersection between AI, SysBio, and SC research over the past two decades, with substantial growth in all three fields and exponential increases in AI-related research in the past decade.

A robust deep learning attack immune MRAM-based physical unclonable function.

The ubiquitous presence of electronic devices demands robust hardware security mechanisms to safeguard sensitive information from threats. This paper presents a physical unclonable function (PUF) circuit based on magnetoresistive random access memory (MRAM). The circuit utilizes inherent characteristics arising from fabrication variations, specifically magnetic tunnel junction (MTJ) cell resistance, to produce corresponding outputs for applied challenges. In contrast to Arbiter PUF, the proposed effectively satisfies the strict avalanche criterion (SAC). Additionally, the grid-like structure of the proposed circuit preserves its resistance against machine learning-based modeling attacks. Various machine learning (ML) attacks employing multilayer perceptron (MLP), linear regression (LR), and support vector machine (SVM) networks are simulated for two-array and four-array architectures. The MLP-attack prediction accuracy was 53.61% for a two-array circuit and 49.87% for a four-array circuit, showcasing robust performance even under the worst-case process variations. In addition, deep learning-based modeling attacks in considerable high dimensions utilizing multiple networks such as convolutional neural network (CNN), recurrent neural network (RNN), MLP, and Larq are used with the accuracy of 50.31%, 50.25%, 50.31%, and 50.31%, respectively. The efficiency of the proposed circuit at the layout level is also investigated for simplified two-array architecture. The simulation results indicate that the proposed circuit offers intra and inter-hamming distance (HD) with a mean of 0.98% and 49.96%, respectively, and a mean diffuseness of 49.09%.

Performance Evaluation of Carrier-Frequency Offset as a Radiometric Fingerprint in Time-Varying Channels.

The authentication of wireless devices through physical layer attributes has attracted a fair amount of attention recently. Recent work in this area has examined various features extracted from the wireless signal to either identify a uniqueness in the channel between the transmitter-receiver pair or more robustly identify certain transmitter behaviors unique to certain devices originating from imperfect hardware manufacturing processes. In particular, the carrier frequency offset (CFO), induced due to the local oscillator mismatch between the transmitter and receiver pair, has exhibited good detection capabilities in stationary and low-mobility transmission scenarios. It is still unclear, however, how the CFO detection capability would hold up in more dynamic time-varying channels where there is a higher mobility. This paper experimentally demonstrates the identification accuracy of CFO for wireless devices in time-varying channels. To this end, a software-defined radio (SDR) testbed is deployed to collect CFO values in real environments, where real transmission and reception are conducted in a vehicular setup. The collected CFO values are used to train machine-learning (ML) classifiers to be used for device identification. While CFO exhibits good detection performance (97% accuracy) for low-mobility scenarios, it is found that higher mobility (35 miles/h) degrades (72% accuracy) the effectiveness of CFO in distinguishing between legitimate and non-legitimate transmitters. This is due to the impact of the time-varying channel on the quality of the exchanged pilot signals used for CFO detection at the receivers.

Effect of urban environment on cardiovascular health: a feasibility pilot study using machine learning to predict heart rate variability in patients with heart failure.

Aims: Urbanization is related to non-communicable diseases such as congestive heart failure (CHF). Understanding the influence of diverse living environments on physiological variables such as heart rate variability (HRV) in patients with chronic cardiac disease may contribute to more effective lifestyle advice and telerehabilitation strategies. This study explores how machine learning (ML) models can predict HRV metrics, which measure autonomic nervous system responses to environmental attributes in uncontrolled real-world settings. The goal is to validate whether this approach can ascertain and quantify the connection between environmental attributes and cardiac autonomic response in patients with CHF. Methods and results: A total of 20 participants (10 healthy individuals and 10 patients with CHF) wore smartwatches for 3 weeks, recording activities, locations, and heart rate (HR). Environmental attributes were extracted from Google Street View images. Machine learning models were trained and tested on the data to predict HRV metrics. The models were evaluated using Spearman's correlation, root mean square error, prediction intervals, and Bland-Altman analysis. Machine learning models predicted HRV metrics related to vagal activity well (R > 0.8 for HR; 0.8 > R > 0.5 for the root mean square of successive interbeat interval differences and the Poincaré plot standard deviation perpendicular to the line of identity; 0.5 > R > 0.4 for the high frequency power and the ratio of the absolute low- and high frequency power induced by environmental attributes. However, they struggled with metrics related to overall autonomic activity, due to the complex balance between sympathetic and parasympathetic modulation. Conclusion: This study highlights the potential of ML-based models to discern vagal dynamics influenced by living environments in healthy individuals and patients diagnosed with CHF. Ultimately, this strategy could offer rehabilitation and tailored lifestyle advice, leading to improved prognosis and enhanced overall patient well-being in CHF.

Rational Design of Lipid Nanoparticles for Enhanced mRNA Vaccine Delivery via Machine Learning.

Since the coronavirus pandemic, mRNA vaccines have revolutionized the field of vaccinology. Lipid nanoparticles (LNPs) are proposed to enhance mRNA delivery efficiency; however, their design is suboptimal. Here, a rational method for designing LNPs is explored, focusing on the ionizable lipid composition and structural optimization using machine learning (ML) techniques. A total of 213 LNPs are analyzed using random forest regression models trained with 314 features to predict the mRNA expression efficiency. The models, which predict mRNA expression levels post-administration of intradermal injection in mice, identify phenol as the dominant substructure affecting mRNA encapsulation and expression. The specific phospholipids used as components of the LNPs, as well as the N/P ratio and mass ratio, are found to affect the efficacy of mRNA delivery. Structural analysis highlights the impact of the carbon chain length on the encapsulation efficiency and LNP stability. This integrated approach offers a framework for designing advanced LNPs and has the potential to unlock the full potential of mRNA therapeutics.

Diagnostic value of an interpretable machine learning model based on clinical ultrasound features for follicular thyroid carcinoma.

Background: Follicular thyroid carcinoma (FTC) and follicular thyroid adenoma (FTA) present diagnostic challenges due to overlapping clinical and ultrasound features. Improving the diagnosis of FTC can enhance patient prognosis and effectiveness in clinical management. This study seeks to develop a predictive model for FTC based on ultrasound features using machine learning (ML) algorithms and assess its diagnostic effectiveness. Methods: Patients diagnosed with FTA or FTC based on surgical pathology between January 2009 and February 2023 at Zhejiang Provincial Cancer Hospital and Zhejiang Provincial People's Hospital were retrospectively included. A total of 562 patients from Zhejiang Provincial Cancer Hospital comprised the training set, and 218 patients from Zhejiang Provincial People's Hospital constituted the validation set. Subsequently, clinical parameters and ultrasound characteristics of the patients were collected. The diagnostic parameters were analyzed using the least absolute shrinkage and selection operator and multivariate logistic regression screening methods. Next, a comparative analysis was performed using seven ML models. The area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, specificity, positive predicted value (PPV), negative predicted value (NPV), precision, recall, and comprehensive evaluation index (F-score) were calculated to compare the diagnostic efficacy among the seven models and determine the optimal model. Further, the optimal model was validated, and the SHapley Additive ExPlanations (SHAP) approach was applied to explain the significance of the model variables. Finally, an individualized risk assessment was conducted. Results: Age, echogenicity, thyroglobulin antibody (TGAb), echotexture, composition, triiodothyronine (T3), thyroglobulin (TG), margin, thyroid-stimulating hormone (TSH), calcification, and halo thickness >2 mm were influential factors for diagnosing FTC. The XGBoost model was identified as the optimal model after a comprehensive evaluation. The AUC of this model in the validation set was 0.969 [95% confidence interval (CI), 0.946-0.992], while its precision sensitivity, specificity, and accuracy were 0.791, 0.930, 0.913 and 0.917, respectively. Conclusions: XGBoost model based on ultrasound features was constructed and interpreted using the SHAP method, providing evidence for the diagnosis of FTC and guidance for the personalized treatment of patients.

Single-Trial Detection and Classification of Event-Related Optical Signals for a Brain-Computer Interface Application.

Event-related optical signals (EROS) measure fast modulations in the brain's optical properties related to neuronal activity. EROS offer a high spatial and temporal resolution and can be used for brain-computer interface (BCI) applications. However, the ability to classify single-trial EROS remains unexplored. This study evaluates the performance of neural network methods for single-trial classification of motor response-related EROS. EROS activity was obtained from a high-density recording montage covering the motor cortex during a two-choice reaction time task involving responses with the left or right hand. This study utilized a convolutional neural network (CNN) approach to extract spatiotemporal features from EROS data and perform classification of left and right motor responses. Subject-specific classifiers trained on EROS phase data outperformed those trained on intensity data, reaching an average single-trial classification accuracy of around 63%. Removing low-frequency noise from intensity data is critical for achieving discriminative classification results with this measure. Our results indicate that deep learning with high-spatial-resolution signals, such as EROS, can be successfully applied to single-trial classifications.