Transfer learning: a friendly introduction

A victim risk identification model for nature-induced urban disaster emergency response

In recent years, nature-induced urban disasters in high-density modern cities in China have raised great concerns. The delayed and imprecise understanding of the real-time post-disaster situation made it difficult for the decision-makers to find a suitable emergency rescue plan. To this end, this study aims to facilitate the real-time performance and accuracy of on-site victim risk identification. In this article, we propose a victim identification model based on the You Only Look Once v7-W6 (YOLOv7-W6) algorithm. This model defines the "fall-down" pose as a key feature in identifying urgent victims from the perspective of disaster medicine rescue. The results demonstrate that this model performs superior accuracy (mAP@0.5, 0.960) and inference speed (5.1 ms) on the established disaster victim database compared to other state-of-the-art object detection algorithms. Finally, a case study is illustrated to show the practical utilization of this model in a real disaster rescue scenario. This study proposes an intelligent on-site victim risk identification approach, contributing significantly to government emergency decision-making and response.

Using AI to Differentiate Mpox From Common Skin Lesions in a Sexual Health Clinic: Algorithm Development and Validation Study

BACKGROUND

The 2022 global outbreak of mpox has significantly impacted health facilities, and necessitated additional infection prevention and control measures and alterations to clinic processes. Early identification of suspected mpox cases will assist in mitigating these impacts.

OBJECTIVE

We aimed to develop and evaluate an artificial intelligence (AI)-based tool to differentiate mpox lesion images from other skin lesions seen in a sexual health clinic.

METHODS

We used a data set with 2200 images, that included mpox and non-mpox lesions images, collected from Melbourne Sexual Health Centre and web resources. We adopted deep learning approaches which involved 6 different deep learning architectures to train our AI models. We subsequently evaluated the performance of each model using a hold-out data set and an external validation data set to determine the optimal model for differentiating between mpox and non-mpox lesions.

RESULTS

The DenseNet-121 model outperformed other models with an overall area under the receiver operating characteristic curve (AUC) of 0.928, an accuracy of 0.848, a precision of 0.942, a recall of 0.742, and an F1-score of 0.834. Implementation of a region of interest approach significantly improved the performance of all models, with the AUC for the DenseNet-121 model increasing to 0.982. This approach resulted in an increase in the correct classification of mpox images from 79% (55/70) to 94% (66/70). The effectiveness of this approach was further validated by a visual analysis with gradient-weighted class activation mapping, demonstrating a reduction in false detection within the background of lesion images. On the external validation data set, ResNet-18 and DenseNet-121 achieved the highest performance. ResNet-18 achieved an AUC of 0.990 and an accuracy of 0.947, and DenseNet-121 achieved an AUC of 0.982 and an accuracy of 0.926.

CONCLUSIONS

Our study demonstrated it was possible to use an AI-based image recognition algorithm to accurately differentiate between mpox and common skin lesions. Our findings provide a foundation for future investigations aimed at refining the algorithm and establishing the place of such technology in a sexual health clinic.

Localized estimation of event-related neural source activity from simultaneous MEG-EEG with a recurrent neural network

Estimating intracranial current sources underlying the electromagnetic signals observed from extracranial sensors is a perennial challenge in non-invasive neuroimaging. Established solutions to this inverse problem treat time samples independently without considering the temporal dynamics of event-related brain processes. This paper describes current source estimation from simultaneously recorded magneto- and electro-encephalography (MEEG) using a recurrent neural network (RNN) that learns sequential relationships from neural data. The RNN was trained in two phases: (1) pre-training and (2) transfer learning with L1 regularization applied to the source estimation layer. Performance of using scaled labels derived from MEEG, magnetoencephalography (MEG), or electroencephalography (EEG) were compared, as were results from volumetric source space with free dipole orientation and surface source space with fixed dipole orientation. Exact low-resolution electromagnetic tomography (eLORETA) and mixed-norm L1/L2 (MxNE) source estimation methods were also applied to these data for comparison with the RNN method. The RNN approach outperformed other methods in terms of output signal-to-noise ratio, correlation and mean-squared error metrics evaluated against reference event-related field (ERF) and event-related potential (ERP) waveforms. Using MEEG labels with fixed-orientation surface sources produced the most consistent estimates. To estimate sources of ERF and ERP waveforms, the RNN generates temporal dynamics within its internal computational units, driven by sequential structure in neural data used as training labels. It thus provides a data-driven model of computational transformations from psychophysiological events into corresponding event-related neural signals, which is unique among MEEG source reconstruction solutions.

The Accurate Prediction of Antibody Deamidations by Combining High-Throughput Automated Peptide Mapping and Protein Language Model-Based Deep Learning

Therapeutic antibodies such as monoclonal antibodies (mAbs), bispecific and multispecific antibodies are pivotal in therapeutic protein development and have transformed disease treatments across various therapeutic areas. The integrity of therapeutic antibodies, however, is compromised by sequence liabilities, notably deamidation, where asparagine (N) and glutamine (Q) residues undergo chemical degradations. Deamidation negatively impacts the efficacy, stability, and safety of diverse classes of antibodies, thus necessitating the critical need for the early and accurate identification of vulnerable sites. In this article, a comprehensive antibody deamidation-specific dataset (n = 2285) of varied modalities was created by using high-throughput automated peptide mapping followed by supervised machine learning to predict the deamidation propensities, as well as the extents, throughout the entire antibody sequences. We propose a novel chimeric deep learning model, integrating protein language model (pLM)-derived embeddings with local sequence information for enhanced deamidation predictions. Remarkably, this model requires only sequence inputs, eliminating the need for laborious feature engineering. Our approach demonstrates state-of-the-art performance, offering a streamlined workflow for high-throughput automated peptide mapping and deamidation prediction, with the potential of broader applicability to other antibody sequence liabilities.

Automatic extraction of transcriptional regulatory interactions of bacteria from biomedical literature using a BERT-based approach

Transcriptional regulatory networks (TRNs) give a global view of the regulatory mechanisms of bacteria to respond to environmental signals. These networks are published in biological databases as a valuable resource for experimental and bioinformatics researchers. Despite the efforts to publish TRNs of diverse bacteria, many of them still lack one and many of the existing TRNs are incomplete. In addition, the manual extraction of information from biomedical literature ("literature curation") has been the traditional way to extract these networks, despite this being demanding and time-consuming. Recently, language models based on pretrained transformers have been used to extract relevant knowledge from biomedical literature. Moreover, the benefit of fine-tuning a large pretrained model with new limited data for a specific task ("transfer learning") opens roads to address new problems of biomedical information extraction. Here, to alleviate this lack of knowledge and assist literature curation, we present a new approach based on the Bidirectional Transformer for Language Understanding (BERT) architecture to classify transcriptional regulatory interactions of bacteria as a first step to extract TRNs from literature. The approach achieved a significant performance in a test dataset of sentences of Escherichia coli (F1-Score: 0.8685, Matthew's correlation coefficient: 0.8163). The examination of model predictions revealed that the model learned different ways to express the regulatory interaction. The approach was evaluated to extract a TRN of Salmonella using 264 complete articles. The evaluation showed that the approach was able to accurately extract 82% of the network and that it was able to extract interactions absent in curation data. To the best of our knowledge, the present study is the first effort to obtain a BERT-based approach to extract this specific kind of interaction. This approach is a starting point to address the limitations of reconstructing TRNs of bacteria and diseases of biological interest. Database URL: https://github.com/laigen-unam/BERT-trn-extraction.

Evaluation of artificial intelligence-powered screening for sexually transmitted infections-related skin lesions using clinical images and metadata

BACKGROUND

Sexually transmitted infections (STIs) pose a significant global public health challenge. Early diagnosis and treatment reduce STI transmission, but rely on recognising symptoms and care-seeking behaviour of the individual. Digital health software that distinguishes STI skin conditions could improve health-seeking behaviour. We developed and evaluated a deep learning model to differentiate STIs from non-STIs based on clinical images and symptoms.

METHODS

We used 4913 clinical images of genital lesions and metadata from the Melbourne Sexual Health Centre collected during 2010-2023. We developed two binary classification models to distinguish STIs from non-STIs: (1) a convolutional neural network (CNN) using images only and (2) an integrated model combining both CNN and fully connected neural network (FCN) using images and metadata. We evaluated the model performance by the area under the ROC curve (AUC) and assessed metadata contributions to the Image-only model.

RESULTS

Our study included 1583 STI and 3330 non-STI images. Common STI diagnoses were syphilis (34.6%), genital warts (24.5%) and herpes (19.4%), while most non-STIs (80.3%) were conditions such as dermatitis, lichen sclerosis and balanitis. In both STI and non-STI groups, the most frequently observed groups were 25-34 years (48.6% and 38.2%, respectively) and heterosexual males (60.3% and 45.9%, respectively). The Image-only model showed a reasonable performance with an AUC of 0.859 (SD 0.013). The Image + Metadata model achieved a significantly higher AUC of 0.893 (SD 0.018) compared to the Image-only model (p < 0.01). Out of 21 metadata, the integration of demographic and dermatological metadata led to the most significant improvement in model performance, increasing AUC by 6.7% compared to the baseline Image-only model.

CONCLUSIONS

The Image + Metadata model outperformed the Image-only model in distinguishing STIs from other skin conditions. Using it as a screening tool in a clinical setting may require further development and evaluation with larger datasets.

The Role of the Neural Exposome as a Novel Strategy to Identify and Mitigate Health Inequities in Alzheimer's Disease and Related Dementias

With the continuous increase of the elderly population, there is an urgency to understand and develop relevant treatments for Alzheimer's disease and related dementias (ADRD). In tandem with this, the prevalence of health inequities continues to rise as disadvantaged communities fail to be included in mainstream research. The neural exposome poses as a relevant mechanistic approach and tool for investigating ADRD onset, progression, and pathology as it accounts for several different factors: exogenous, endogenous, and behavioral. Consequently, through the neural exposome, health inequities can be addressed in ADRD research. In this paper, we address how the neural exposome relates to ADRD by contributing to the discourse through defining how the neural exposome can be developed as a tool in accordance with machine learning. Through this, machine learning can allow for developing a greater insight into the application of transferring and making sense of experimental mouse models exposed to health inequities and potentially relate it to humans. The overall goal moving beyond this paper is to define a multitude of potential factors that can increase the risk of ADRD onset and integrate them to create an interdisciplinary approach to the study of ADRD and subsequently translate the findings to clinical research.

Role of Artificial Intelligence in Multinomial Decisions and Preventative Nutrition in Alzheimer's Disease

Alzheimer's disease (AD) affects 50 million people worldwide, an increase of 35 million since 2015, and it is known for memory loss and cognitive decline. Considering the morbidity associated with AD, it is important to explore lifestyle elements influencing the chances of developing AD, with special emphasis on nutritional aspects. This review will first discuss how dietary factors have an impact in AD development and the possible role of Artificial Intelligence (AI) and Machine Learning (ML) in preventative care of AD patients through nutrition. The Mediterranean-DASH diets provide individuals with many nutrient benefits which assists the prevention of neurodegeneration by having neuroprotective roles. Lack of micronutrients, protein-energy, and polyunsaturated fatty acids increase the chance of cognitive decline, loss of memory, and synaptic dysfunction among others. ML software has the ability to design models of algorithms from data introduced to present practical solutions that are accessible and easy to use. It can give predictions for a precise medicine approach to evaluate individuals as a whole. There is no doubt the future of nutritional science lies on customizing diets for individuals to reduce dementia risk factors, maintain overall health and brain function.

Transfer Learning in Cancer Genetics, Mutation Detection, Gene Expression Analysis, and Syndrome Recognition

Artificial intelligence (AI), encompassing machine learning (ML) and deep learning (DL), has revolutionized medical research, facilitating advancements in drug discovery and cancer diagnosis. ML identifies patterns in data, while DL employs neural networks for intricate processing. Predictive modeling challenges, such as data labeling, are addressed by transfer learning (TL), leveraging pre-existing models for faster training. TL shows potential in genetic research, improving tasks like gene expression analysis, mutation detection, genetic syndrome recognition, and genotype-phenotype association. This review explores the role of TL in overcoming challenges in mutation detection, genetic syndrome detection, gene expression, or phenotype-genotype association. TL has shown effectiveness in various aspects of genetic research. TL enhances the accuracy and efficiency of mutation detection, aiding in the identification of genetic abnormalities. TL can improve the diagnostic accuracy of syndrome-related genetic patterns. Moreover, TL plays a crucial role in gene expression analysis in order to accurately predict gene expression levels and their interactions. Additionally, TL enhances phenotype-genotype association studies by leveraging pre-trained models. In conclusion, TL enhances AI efficiency by improving mutation prediction, gene expression analysis, and genetic syndrome detection. Future studies should focus on increasing domain similarities, expanding databases, and incorporating clinical data for better predictions.

Investigating the detection of breast cancer with deep transfer learning using ResNet18 and ResNet34

A lot of underdeveloped nations particularly in Africa struggle with cancer-related, deadly diseases. Particularly in women, the incidence of breast cancer is rising daily because of ignorance and delayed diagnosis. Only by correctly identifying and diagnosing cancer in its very early stages of development can be effectively treated. The classification of cancer can be accelerated and automated with the aid of computer-aided diagnosis and medical image analysis techniques. This research provides the use of transfer learning from a Residual Network 18 (ResNet18) and Residual Network 34 (ResNet34) architectures to detect breast cancer. The study examined how breast cancer can be identified in breast mammography pictures using transfer learning from ResNet18 and ResNet34, and developed a demo app for radiologists using the trained models with the best validation accuracy. 1, 200 datasets of breast x-ray mammography images from the National Radiological Society's (NRS) archives were employed in the study. The dataset was categorised as implant cancer negative, implant cancer positive, cancer negative and cancer positive in order to increase the consistency of x-ray mammography images classification and produce better features. For the multi-class classification of the images, the study gave an average accuracy for binary classification of benign or malignant cancer cases of 86.7% validation accuracy for ResNet34 and 92% validation accuracy for ResNet18. A prototype web application showcasing ResNet18 performance has been created. The acquired results show how transfer learning can improve the accuracy of breast cancer detection, providing invaluable assistance to medical professionals, particularly in an African scenario.

A Review for Artificial Intelligence Based Protein Subcellular Localization

Proteins need to be located in appropriate spatiotemporal contexts to carry out their diverse biological functions. Mislocalized proteins may lead to a broad range of diseases, such as cancer and Alzheimer's disease. Knowing where a target protein resides within a cell will give insights into tailored drug design for a disease. As the gold validation standard, the conventional wet lab uses fluorescent microscopy imaging, immunoelectron microscopy, and fluorescent biomarker tags for protein subcellular location identification. However, the booming era of proteomics and high-throughput sequencing generates tons of newly discovered proteins, making protein subcellular localization by wet-lab experiments a mission impossible. To tackle this concern, in the past decades, artificial intelligence (AI) and machine learning (ML), especially deep learning methods, have made significant progress in this research area. In this article, we review the latest advances in AI-based method development in three typical types of approaches, including sequence-based, knowledge-based, and image-based methods. We also elaborately discuss existing challenges and future directions in AI-based method development in this research field.

Offloading the computational complexity of transfer learning with generic features

Deep learning approaches are generally complex, requiring extensive computational resources and having high time complexity. Transfer learning is a state-of-the-art approach to reducing the requirements of high computational resources by using pre-trained models without compromising accuracy and performance. In conventional studies, pre-trained models are trained on datasets from different but similar domains with many domain-specific features. The computational requirements of transfer learning are directly dependent on the number of features that include the domain-specific and the generic features. This article investigates the prospects of reducing the computational requirements of the transfer learning models by discarding domain-specific features from a pre-trained model. The approach is applied to breast cancer detection using the dataset curated breast imaging subset of the digital database for screening mammography and various performance metrics such as precision, accuracy, recall, F1-score, and computational requirements. It is seen that discarding the domain-specific features to a specific limit provides significant performance improvements as well as minimizes the computational requirements in terms of training time (reduced by approx. 12%), processor utilization (reduced approx. 25%), and memory usage (reduced approx. 22%). The proposed transfer learning strategy increases accuracy (approx. 7%) and offloads computational complexity expeditiously.

Adjustment of scRNA-seq data to improve cell-type decomposition of spatial transcriptomics

Most sequencing-based spatial transcriptomics (ST) technologies do not achieve single-cell resolution where each captured location (spot) may contain a mixture of cells from heterogeneous cell types, and several cell-type decomposition methods have been proposed to estimate cell type proportions of each spot by integrating with single-cell RNA sequencing (scRNA-seq) data. However, these existing methods did not fully consider the effect of distribution difference between scRNA-seq and ST data for decomposition, leading to biased cell-type-specific genes derived from scRNA-seq for ST data. To address this issue, we develop an instance-based transfer learning framework to adjust scRNA-seq data by ST data to correctly match cell-type-specific gene expression. We evaluate the effect of raw and adjusted scRNA-seq data on cell-type decomposition by eight leading decomposition methods using both simulated and real datasets. Experimental results show that data adjustment can effectively reduce distribution difference and improve decomposition, thus enabling for a more precise depiction on spatial organization of cell types. We highlight the importance of data adjustment in integrative analysis of scRNA-seq with ST data and provide guidance for improved cell-type decomposition.

Assessing the Utility of artificial intelligence in endometriosis: Promises and pitfalls

Endometriosis, a chronic condition characterized by the growth of endometrial-like tissue outside of the uterus, poses substantial challenges in terms of diagnosis and treatment. Artificial intelligence (AI) has emerged as a promising tool in the field of medicine, offering opportunities to address the complexities of endometriosis. This review explores the current landscape of endometriosis diagnosis and treatment, highlighting the potential of AI to alleviate some of the associated burdens and underscoring common pitfalls and challenges when employing AI algorithms in this context. Women's health research in endometriosis has suffered from underfunding, leading to limitations in diagnosis, classification, and treatment approaches. The heterogeneity of symptoms in patients with endometriosis has further complicated efforts to address this condition. New, powerful methods of analysis have the potential to uncover previously unidentified patterns in data relating to endometriosis. AI, a collection of algorithms replicating human decision-making in data analysis, has been increasingly adopted in medical research, including endometriosis studies. While AI offers the ability to identify novel patterns in data and analyze large datasets, its effectiveness hinges on data quality and quantity and the expertise of those implementing the algorithms. Current applications of AI in endometriosis range from diagnostic tools for ultrasound imaging to predicting treatment success. These applications show promise in reducing diagnostic delays, healthcare costs, and providing patients with more treatment options, improving their quality of life. AI holds significant potential in advancing the diagnosis and treatment of endometriosis, but it must be applied carefully and transparently to avoid pitfalls and ensure reproducibility. This review calls for increased scrutiny and accountability in AI research. Addressing these challenges can lead to more effective AI-driven solutions for endometriosis and other complex medical conditions.

A rapid and reference-free imputation method for low-cost genotyping platforms

Most current genotype imputation methods are reference-based, which posed several challenges to users, such as high computational costs and reference panel inaccessibility. Thus, deep learning models are expected to create reference-free imputation methods performing with higher accuracy and shortening the running time. We proposed a imputation method using recurrent neural networks integrating with an additional discriminator network, namely GRUD. This method was applied to datasets from genotyping chips and Low-Pass Whole Genome Sequencing (LP-WGS) with the reference panels from The 1000 Genomes Project (1KGP) phase 3, the dataset of 4810 Singaporeans (SG10K), and The 1000 Vietnamese Genome Project (VN1K). Our model performed more accurately than other existing methods on multiple datasets, especially with common variants with large minor allele frequency, and shrank running time and memory usage. In summary, these results indicated that GRUD can be implemented in genomic analyses to improve the accuracy and running-time of genotype imputation.

Transferable Implicit Solvation via Contrastive Learning of Graph Neural Networks

Implicit solvent models are essential for molecular dynamics simulations of biomolecules, striking a balance between computational efficiency and biological realism. Efforts are underway to develop accurate and transferable implicit solvent models and coarse-grained (CG) force fields in general, guided by a bottom-up approach that matches the CG energy function with the potential of mean force (PMF) defined by the finer system. However, practical challenges arise due to the lack of analytical expressions for the PMF and algorithmic limitations in parameterizing CG force fields. To address these challenges, a machine learning-based approach is proposed, utilizing graph neural networks (GNNs) to represent the solvation free energy and potential contrasting for parameter optimization. We demonstrate the effectiveness of the approach by deriving a transferable GNN implicit solvent model using 600,000 atomistic configurations of six proteins obtained from explicit solvent simulations. The GNN model provides solvation free energy estimations much more accurately than state-of-the-art implicit solvent models, reproducing configurational distributions of explicit solvent simulations. We also demonstrate the reasonable transferability of the GNN model outside of the training data. Our study offers valuable insights for deriving systematically improvable implicit solvent models and CG force fields from a bottom-up perspective.

Survey of Transfer Learning Approaches in the Machine Learning of Digital Health Sensing Data

Machine learning and digital health sensing data have led to numerous research achievements aimed at improving digital health technology. However, using machine learning in digital health poses challenges related to data availability, such as incomplete, unstructured, and fragmented data, as well as issues related to data privacy, security, and data format standardization. Furthermore, there is a risk of bias and discrimination in machine learning models. Thus, developing an accurate prediction model from scratch can be an expensive and complicated task that often requires extensive experiments and complex computations. Transfer learning methods have emerged as a feasible solution to address these issues by transferring knowledge from a previously trained task to develop high-performance prediction models for a new task. This survey paper provides a comprehensive study of the effectiveness of transfer learning for digital health applications to enhance the accuracy and efficiency of diagnoses and prognoses, as well as to improve healthcare services. The first part of this survey paper presents and discusses the most common digital health sensing technologies as valuable data resources for machine learning applications, including transfer learning. The second part discusses the meaning of transfer learning, clarifying the categories and types of knowledge transfer. It also explains transfer learning methods and strategies, and their role in addressing the challenges in developing accurate machine learning models, specifically on digital health sensing data. These methods include feature extraction, fine-tuning, domain adaptation, multitask learning, federated learning, and few-/single-/zero-shot learning. This survey paper highlights the key features of each transfer learning method and strategy, and discusses the limitations and challenges of using transfer learning for digital health applications. Overall, this paper is a comprehensive survey of transfer learning methods on digital health sensing data which aims to inspire researchers to gain knowledge of transfer learning approaches and their applications in digital health, enhance the current transfer learning approaches in digital health, develop new transfer learning strategies to overcome the current limitations, and apply them to a variety of digital health technologies.

A Systematic Review of Machine-Learning Solutions in Anaerobic Digestion

The use of machine learning (ML) in anaerobic digestion (AD) is growing in popularity and improves the interpretation of complex system parameters for better operation and optimisation. This systematic literature review aims to explore how ML is currently employed in AD, with particular attention to the challenges of implementation and the benefits of integrating ML techniques. While both lab and industry-scale datasets have been used for model training, challenges arise from varied system designs and the different monitoring equipment used. Traditional machine-learning techniques, predominantly artificial neural networks (ANN), are the most commonly used but face difficulties in scalability and interpretability. Specifically, models trained on lab-scale data often struggle to generalize to full-scale, real-world operations due to the complexity and variability in bacterial communities and system operations. In practical scenarios, machine learning can be employed in real-time operations for predictive modelling, ensuring system stability is maintained, resulting in improved efficiency of both biogas production and waste treatment processes. Through reviewing the ML techniques employed in wider applied domains, potential future research opportunities in addressing these challenges have been identified.

Diagnosis of Salivary Gland Tumors Using Transfer Learning with Fine-Tuning and Gradual Unfreezing

Ultrasound is the primary tool for evaluating salivary gland tumors (SGTs); however, tumor diagnosis currently relies on subjective features. This study aimed to establish an objective ultrasound diagnostic method using deep learning. We collected 446 benign and 223 malignant SGT ultrasound images in the training/validation set and 119 benign and 44 malignant SGT ultrasound images in the testing set. We trained convolutional neural network (CNN) models from scratch and employed transfer learning (TL) with fine-tuning and gradual unfreezing to classify malignant and benign SGTs. The diagnostic performances of these models were compared. By utilizing the pretrained ResNet50V2 with fine-tuning and gradual unfreezing, we achieved a 5-fold average validation accuracy of 0.920. The diagnostic performance on the testing set demonstrated an accuracy of 89.0%, a sensitivity of 81.8%, a specificity of 91.6%, a positive predictive value of 78.3%, and a negative predictive value of 93.2%. This performance surpasses that of other models in our study. The corresponding Grad-CAM visualizations were also presented to provide explanations for the diagnosis. This study presents an effective and objective ultrasound method for distinguishing between malignant and benign SGTs, which could assist in preoperative evaluation.

Bayesian parametric models for survival prediction in medical applications

BACKGROUND

Evidence-based treatment decisions in medicine are made founded on population-level evidence obtained during randomized clinical trials. In an era of personalized medicine, these decisions should be based on the predicted benefit of a treatment on a patient-level. Survival prediction models play a central role as they incorporate the time-to-event and censoring. In medical applications uncertainty is critical especially when treatments differ in their side effect profiles or costs. Additionally, models must be adapted to local populations without diminishing performance and often without the original training data available due to privacy concern. Both points are supported by Bayesian models-yet they are rarely used. The aim of this work is to evaluate Bayesian parametric survival models on public datasets including cardiology, infectious diseases, and oncology.

MATERIALS AND METHODS

Bayesian parametric survival models based on the Exponential and Weibull distribution were implemented as a Python package. A linear combination and a neural network were used for predicting the parameters of the distributions. A superiority design was used to assess whether Bayesian models are better than commonly used models such as Cox Proportional Hazards, Random Survival Forest, and Neural Network-based Cox Proportional Hazards. In a secondary analysis, overfitting was compared between these models. An equivalence design was used to assess whether the prediction performance of Bayesian models after model updating using Bayes rule is equivalent to retraining on the full dataset.

RESULTS

In this study, we found that Bayesian parametric survival models perform as good as state-of-the art models while requiring less hyperparameters to be tuned and providing a measure of the uncertainty of the predictions. In addition, these models were less prone to overfitting. Furthermore, we show that updating these models using Bayes rule yields equivalent performance compared to models trained on combined original and new datasets.

CONCLUSIONS

Bayesian parametric survival models are non-inferior to conventional survival models while requiring less hyperparameter tuning, being less prone to overfitting, and allowing model updating using Bayes rule. Further, the Bayesian models provide a measure of the uncertainty on the statistical inference, and, in particular, on the prediction.

Machine learning analysis and risk prediction of weather-sensitive mortality related to cardiovascular disease during summer in Tokyo, Japan

Climate-sensitive diseases developing from heat or cold stress threaten human health. Therefore, the future health risk induced by climate change and the aging of society need to be assessed. We developed a prediction model for mortality due to cardiovascular diseases such as myocardial infarction and cerebral infarction, which are weather or climate sensitive, using machine learning (ML) techniques. We evaluated the daily mortality of ischaemic heart disease (IHD) and cerebrovascular disease (CEV) in Tokyo and Osaka City, Japan, during summer. The significance of delayed effects of daily maximum temperature and other weather elements on mortality was previously demonstrated using a distributed lag nonlinear model. We conducted ML by a LightGBM algorithm that included specified lag days, with several temperature- and air pressure-related elements, to assess the respective mortality risks for IHD and CEV, based on training and test data for summer 2010-2019. These models were used to evaluate the effect of climate change on the risk for IHD mortality in Tokyo by applying transfer learning (TL). ML with TL predicted that the daily IHD mortality risk in Tokyo would averagely increase by 29% and 35% at the 95th and 99th percentiles, respectively, using a high-level warming-climate scenario in 2045-2055, compared to the risk simulated using ML in 2009-2019.

RANet: a custom CNN model and quanvolutional neural network for the automated detection of rheumatoid arthritis in hand thermal images

Rheumatoid arthritis is an autoimmune disease which affects the small joints. Early prediction of RA is necessary for the treatment and management of the disease. The current work presents a deep learning and quantum computing-based automated diagnostic approach for RA in hand thermal imaging. The study's goals are (i) to develop a custom RANet model and compare its performance with the pretrained models and quanvolutional neural network (QNN) to distinguish between the healthy subjects and RA patients, (ii) To validate the performance of the custom model using feature selection method and classification using machine learning (ML) classifiers. The present study developed a custom RANet model and employed pre-trained models such as ResNet101V2, InceptionResNetV2, and DenseNet201 to classify the RA patients and normal subjects. The deep features extracted from the RA Net model are fed into the ML classifiers after the feature selection process. The RANet model, RA Net+ SVM, and QNN model produced an accuracy of 95%, 97% and 93.33% respectively in the classification of healthy groups and RA patients. The developed RANet and QNN models based on thermal imaging could be employed as an accurate automated diagnostic tool to differentiate between the RA and control groups.

Transferable Coarse Graining via Contrastive Learning of Graph Neural Networks

Coarse-grained (CG) force fields are essential for molecular dynamics simulations of biomolecules, striking a balance between computational efficiency and biological realism. These simulations employ simplified models grouping atoms into interaction sites, enabling the study of complex biomolecular systems over biologically relevant timescales. Efforts are underway to develop accurate and transferable CG force fields, guided by a bottom-up approach that matches the CG energy function with the potential of mean force (PMF) defined by the finer system. However, practical challenges arise due to many-body effects, lack of analytical expressions for the PMF, and limitations in parameterizing CG force fields. To address these challenges, a machine learning-based approach is proposed, utilizing graph neural networks (GNNs) to represent CG force fields and potential contrasting for parameterization from atomistic simulation data. We demonstrate the effectiveness of the approach by deriving a transferable GNN implicit solvent model using 600,000 atomistic configurations of six proteins obtained from explicit solvent simulations. The GNN model provides solvation free energy estimations much more accurately than state-of-the-art implicit solvent models, reproducing configurational distributions of explicit solvent simulations. We also demonstrate the reasonable transferability of the GNN model outside the training data. Our study offers valuable insights for building accurate coarse-grained models bottom-up.

Cross-Domain Indoor Visual Place Recognition for Mobile Robot via Generalization Using Style Augmentation

The article presents an algorithm for the multi-domain visual recognition of an indoor place. It is based on a convolutional neural network and style randomization. The authors proposed a scene classification mechanism and improved the performance of the models based on synthetic and real data from various domains. In the proposed dataset, a domain change was defined as a camera model change. A dataset of images collected from several rooms was used to show different scenarios, human actions, equipment changes, and lighting conditions. The proposed method was tested in a scene classification problem where multi-domain data were used. The basis was a transfer learning approach with an extension style applied to various combinations of source and target data. The focus was on improving the unknown domain score and multi-domain support. The results of the experiments were analyzed in the context of data collected on a humanoid robot. The article shows that the average score was the highest for the use of multi-domain data and data style enhancement. The method of obtaining average results for the proposed method reached the level of 92.08%. The result obtained by another research team was corrected.

Emotion Recognition from Spatio-Temporal Representation of EEG Signals via 3D-CNN with Ensemble Learning Techniques

The recognition of emotions is one of the most challenging issues in human-computer interaction (HCI). EEG signals are widely adopted as a method for recognizing emotions because of their ease of acquisition, mobility, and convenience. Deep neural networks (DNN) have provided excellent results in emotion recognition studies. Most studies, however, use other methods to extract handcrafted features, such as Pearson correlation coefficient (PCC), Principal Component Analysis, Higuchi Fractal Dimension (HFD), etc., even though DNN is capable of generating meaningful features. Furthermore, most earlier studies largely ignored spatial information between the different channels, focusing mainly on time domain and frequency domain representations. This study utilizes a pre-trained 3D-CNN MobileNet model with transfer learning on the spatio-temporal representation of EEG signals to extract features for emotion recognition. In addition to fully connected layers, hybrid models were explored using other decision layers such as multilayer perceptron (MLP), k-nearest neighbor (KNN), extreme learning machine (ELM), XGBoost (XGB), random forest (RF), and support vector machine (SVM). Additionally, this study investigates the effects of post-processing or filtering output labels. Extensive experiments were conducted on the SJTU Emotion EEG Dataset (SEED) (three classes) and SEED-IV (four classes) datasets, and the results obtained were comparable to the state-of-the-art. Based on the conventional 3D-CNN with ELM classifier, SEED and SEED-IV datasets showed a maximum accuracy of 89.18% and 81.60%, respectively. Post-filtering improved the emotional classification performance in the hybrid 3D-CNN with ELM model for SEED and SEED-IV datasets to 90.85% and 83.71%, respectively. Accordingly, spatial-temporal features extracted from the EEG, along with ensemble classifiers, were found to be the most effective in recognizing emotions compared to state-of-the-art methods.