J Am Coll Cardiol. 2024 Aug 27;84(9):815-828. doi: 10.1016/j.jacc.2024.05.062.

ABSTRACT

BACKGROUND: Artificial intelligence-enhanced electrocardiogram (AI-ECG) analysis shows promise to detect biventricular pathophysiology. However, AI-ECG analysis remains underexplored in congenital heart disease (CHD).

OBJECTIVES: The purpose of this study was to develop and externally validate an AI-ECG model to predict cardiovascular magnetic resonance (CMR)-defined biventricular dysfunction/dilation in patients with CHD.

METHODS: We trained (80%) and tested (20%) a convolutional neural network on paired ECG-CMRs (≤30 days apart) from patients with and without CHD to detect left ventricular (LV) dysfunction (ejection fraction ≤40%), RV dysfunction (ejection fraction ≤35%), and LV and RV dilation (end-diastolic volume z-score ≥4). Performance was assessed during internal testing and external validation on an outside health care system using area under receiver-operating curve (AUROC) and area under precision recall curve.

RESULTS: The internal and external cohorts comprised 8,584 ECG-CMR pairs (n = 4,941; median CMR age 20.7 years) and 909 ECG-CMR pairs (n = 746; median CMR age 25.4 years), respectively. Model performance was similar for internal testing (AUROC: LV dysfunction 0.87; LV dilation 0.86; RV dysfunction 0.88; RV dilation 0.81) and external validation (AUROC: LV dysfunction 0.89; LV dilation 0.83; RV dysfunction 0.82; RV dilation 0.80). Model performance was lowest in functionally single ventricle patients. Tetralogy of Fallot patients predicted to be at high risk of ventricular dysfunction had lower survival (P < 0.001). Model explainability via saliency mapping revealed that lateral precordial leads influence all outcome predictions, with high-risk features including QRS widening and T-wave inversions for RV dysfunction/dilation.

CONCLUSIONS: AI-ECG shows promise to predict biventricular dysfunction/dilation, which may help inform CMR timing in CHD.

PMID:39168568 | DOI:10.1016/j.jacc.2024.05.062

Heart Rhythm. 2024 Aug 19:S1547-5271(24)03149-7. doi: 10.1016/j.hrthm.2024.08.030. Online ahead of print.

ABSTRACT

BACKGROUND: Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a rare genetic heart disease associated with life-threatening ventricular arrhythmias. Diagnosis of ARVC is based on the 2010 Task Force Criteria (TFC), application of which often requires clinical expertise at specialized centers.

OBJECTIVE: To develop and validate an electrocardiogram (ECG) deep-learning (DL) tool for ARVC diagnosis.

METHODS: ECGs of patients referred for ARVC evaluation were used to develop (n=551, 80.1%) and test (n=137, 19.9%) an ECG-DL model for prediction of TFC-defined ARVC diagnosis. The ARVC ECG-DL model was externally validated in a cohort of patients with pathogenic or likely pathogenic (P/LP) ARVC gene variants identified through the Geisinger MyCode Community Health Initiative (N=167).

RESULTS: Of 688 patients evaluated at JHH (57.3% male, mean age 40.2 years), 329 (47.8%) were diagnosed with ARVC. While ARVC diagnosis made by referring cardiologist ECG interpretation was unreliable (c-statistic 0.53 [CI: 0.52, 0.53]), ECG-DL discrimination in the hold-out testing cohort was excellent (0.87 [0.86, 0.89]) and compared favorably to that of ECG interpretation by an ARVC expert (0.85 [0.84, 0.86]). In the Geisinger cohort, prevalence of ARVC was lower (n=17, 10.2%), but ECG-DL based identification of ARVC phenotype remained reliable (0.80 [0.77, 0.83]). Discrimination was further increased when ECG-DL predictions were combined with non-ECG-derived TFC in the JHH testing (c-statistic 0.940 [95%CI: 0.933; 0.948]) and Geisinger validation (0.897 [95%CI: 0.883; 0.912]) cohorts.

CONCLUSION: ECG-DL augments diagnosis of ARVC to the level of an ARVC expert and can differentiate true ARVC diagnosis from phenotype-mimics and at-risk family members/genotype-positive individuals.

PMID:39168295 | DOI:10.1016/j.hrthm.2024.08.030

Methods. 2024 Aug 19:S1046-2023(24)00185-3. doi: 10.1016/j.ymeth.2024.08.005. Online ahead of print.

ABSTRACT

Promoters, which are short (50-1500 base-pair) in DNA regions, have emerged to play a critical role in the regulation of gene transcription. Numerous dangerous diseases, likewise cancer, cardiovascular, and inflammatory bowel diseases, are caused by genetic variations in promoters. Consequently, the correct identification and characterization of promoters are significant for the discovery of drugs. However, experimental approaches to recognizing promoters and their strengths are challenging in terms of cost, time, and resources. Therefore, computational techniques are highly desirable for the correct characterization of promoters from unannotated genomic data. Here, we designed a powerful bi-layer deep-learning based predictor named "PROCABLES", which discriminates DNA samples as promoters in the first-phase and strong or weak promoters in the second-phase respectively. The proposed method utilizes five distinct features, such as word2vec, k-spaced nucleotide pairs, trinucleotide propensity-based features, trinucleotide composition, and electron-ion interaction pseudopotentials, to extract the hidden patterns from the DNA sequence. Afterwards, a stacked framework is formed by integrating a convolutional neural network (CNN) with bidirectional long-short-term memory (LSTM) using multi-view attributes to train the proposed model. The PROCABLES model achieved an accuracy of 0.971 and 0.920 and the MCC 0.940 and 0.840 for the first and second-layer using the ten-fold cross-validation test, respectively. The predicted results anticipate that the proposed PROCABLES protocol outperformed the advanced computational predictors targeting promoters and their types. In summary, this research will provide useful hints for the recognition of large-scale promoters in particular and other DNA problems in general.

PMID:39168294 | DOI:10.1016/j.ymeth.2024.08.005

J Neurosci Methods. 2024 Aug 19:110253. doi: 10.1016/j.jneumeth.2024.110253. Online ahead of print.

ABSTRACT

BACKGROUND: There is a broad interest in deploying deep learning-based classification algorithms to identify individuals with Alzheimer's disease (AD) from healthy controls (HC) based on neuroimaging data, such as T1-weighted Magnetic Resonance Imaging (MRI). The goal of the current study is to investigate whether modern, flexible architectures such as EfficientNet provide any performance boost over more standard architectures.

METHODS: MRI data was sourced from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and processed with a minimal preprocessing pipeline. Among the various architectures tested, the minimal 3D convolutional neural network SFCN stood out, composed solely of 3x3x3 convolution, batch normalization, ReLU, and max-pooling. We also examined the influence of scale on performance, testing SFCN versions with trainable parameters ranging from 720 up to 2.9 million.

RESULTS: SFCN achieves a test ROC AUC of 96.0% while EfficientNet got an ROC AUC of 94.9 %. SFCN retained high performance down to 720 trainable parameters, achieving an ROC AUC of 91.4%.

COMPARISON WITH EXISTING METHODS: The SFCN is compared to DenseNet and EfficientNet as well as the results of other publications in the field.

CONCLUSIONS: The results indicate that using the minimal 3D convolutional neural network SFCN with a minimal preprocessing pipeline can achieve competitive performance in AD classification, challenging the necessity of employing more complex architectures with a larger number of parameters. This finding supports the efficiency of simpler deep learning models for neuroimaging-based AD diagnosis, potentially aiding in better understanding and diagnosing Alzheimer's disease.

PMID:39168252 | DOI:10.1016/j.jneumeth.2024.110253

Med Image Anal. 2024 Aug 19;98:103305. doi: 10.1016/j.media.2024.103305. Online ahead of print.

ABSTRACT

Three-dimensional (3D) freehand ultrasound (US) is a widely used imaging modality that allows non-invasive imaging of medical anatomy without radiation exposure. Surface reconstruction of US volume is vital to acquire the accurate anatomical structures needed for modeling, registration, and visualization. However, traditional methods cannot produce a high-quality surface due to image noise. Despite improvements in smoothness, continuity, and resolution from deep learning approaches, research on surface reconstruction in freehand 3D US is still limited. This study introduces FUNSR, a self-supervised neural implicit surface reconstruction method to learn signed distance functions (SDFs) from US volumes. In particular, FUNSR iteratively learns the SDFs by moving the 3D queries sampled around volumetric point clouds to approximate the surface, guided by two novel geometric constraints: sign consistency constraint and on-surface constraint with adversarial learning. Our approach has been thoroughly evaluated across four datasets to demonstrate its adaptability to various anatomical structures, including a hip phantom dataset, two vascular datasets and one publicly available prostate dataset. We also show that smooth and continuous representations greatly enhance the visual appearance of US data. Furthermore, we highlight the potential of our method to improve segmentation performance, and its robustness to noise distribution and motion perturbation.

PMID:39168075 | DOI:10.1016/j.media.2024.103305

Neural Netw. 2024 Aug 3;179:106592. doi: 10.1016/j.neunet.2024.106592. Online ahead of print.

ABSTRACT

Brain age (BA) is defined as a measure of brain maturity and could help characterize both the typical brain development and neuropsychiatric disorders in mammals. Various biological phenotypes have been successfully applied to predict BA of human using chronological age (CA) as label. However, whether the BA of macaque, one of the most important animal models, can also be reliably predicted is largely unknown. To address this question, we propose a novel deep learning model called Multi-Branch Vision Transformer (MB-ViT) to fuse multi-scale (i.e., from coarse-grained to fine-grained) brain functional connectivity (FC) patterns derived from resting state functional magnetic resonance imaging (rs-fMRI) data to predict BA of macaques. The discriminative functional connections and the related brain regions contributing to the prediction are further identified based on Gradient-weighted Class Activation Mapping (Grad-CAM) method. Our proposed model successfully predicts BA of 450 normal rhesus macaques from the publicly available PRIMatE Data Exchange (PRIME-DE) dataset with lower mean absolute error (MAE) and mean square error (MSE) as well as higher Pearson's correlation coefficient (PCC) and coefficient of determination (R2) compared to other baseline models. The correlation between the predicted BA and CA reaches as high as 0.82 of our proposed method. Furthermore, our analysis reveals that the functional connections predominantly contributing to the prediction results are situated in the primary motor cortex (M1), visual cortex, area v23 in the posterior cingulate cortex, and dysgranular temporal pole. In summary, our proposed deep learning model provides an effective tool to accurately predict BA of primates (macaque in this study), and lays a solid foundation for future studies of age-related brain diseases in those animal models.

PMID:39168070 | DOI:10.1016/j.neunet.2024.106592

Talanta. 2024 Aug 8;280:126693. doi: 10.1016/j.talanta.2024.126693. Online ahead of print.

ABSTRACT

Perfluorooctanoic acid (PFOA) has received increasing concerns in recent years due to its wide distribution and potential toxicity. Existing detection techniques of PFOA require complex pre-treatment, therefore often taking several hours. Here, we developed a rapid PFOA detection mode to detect approximate concentrations of PFOA (ranging from 10-15 to 10-3 mol/L) in deionized water, and detecting one sample takes only 20 min. The detection mode was achieved using a deep learning model trained by a large surface enhanced Raman spectra dataset, based on the agglomeration of PFOA with crystal violet. In addition, transfer learning approach was used to fine tune the model, the fine-tuned model was generalizable across water samples with different impurities and environments to determine whether meet the safety standards of PFOA, the accuracy was 96.25 % and 94.67 % for tap water and lake water samples, respectively. The mechanism and specificity of the detection mode were further confirmed by molecular dynamics simulation. Our work provides a promising solution for PFOA detection, especially in the context of the increasingly widespread application of PFOA.

PMID:39167934 | DOI:10.1016/j.talanta.2024.126693

Nurse Educ Today. 2024 Aug 14;142:106356. doi: 10.1016/j.nedt.2024.106356. Online ahead of print.

ABSTRACT

BACKGROUND: Adequate professional preparation of nursing undergraduates is conducive to developing health care careers. Deep learning is important for enhancing nursing competencies and the overall quality of students. However, limited research has been conducted to explore deep learning and its associated factors for students in higher nursing education.

OBJECTIVE: To describe the level of deep learning and explore its associated factors among Chinese nursing undergraduates.

DESIGN: A cross-sectional study.

SETTING: This study was conducted at a medical university in Anhui Province, China.

PARTICIPANTS: Convenience sampling was used to survey 271 nursing undergraduates between July and September 2023.

METHODS: The survey included questions about general information, deep learning, and critical thinking disposition. Nonparametric tests were used to distinguish the intergroup differences. Correlations were evaluated using Spearman's rank correlation analysis. Hierarchical multiple regression analysis was performed to determine the influencing factors.

RESULTS: The deep learning score of the nursing undergraduates was 3.82 (3.56, 4.00). Hierarchical multiple regression analysis revealed that gender (β = 0.10, P = 0.044), experience as a student leader (β = 0.10, P = 0.049), and critical thinking disposition (β = 0.60, P = 0.000) significantly impacted deep learning. All the variables explained 41.1 % of the total mean score variance for deep learning.

CONCLUSION: Chinese nursing undergraduates showed upper-middle levels of deep learning. Gender, experience as a student leader, and critical thinking disposition were significantly associated factors of deep learning. Nursing educators should provide targeted interventions for deep learning to facilitate the professional competencies of these students.

PMID:39167874 | DOI:10.1016/j.nedt.2024.106356

J Orthop Surg Res. 2024 Aug 21;19(1):496. doi: 10.1186/s13018-024-05002-5.

ABSTRACT

BACKGROUND: In recent years, deep learning (DL) technology has been increasingly used for the diagnosis and treatment of lumbar intervertebral disc (IVD) degeneration. This study aims to evaluate the performance of DL technology for IVD segmentation in magnetic resonance (MR) images and explore improvement strategies.

METHODS: We developed a PRISMA systematic review protocol and systematically reviewed studies that used DL algorithm frameworks to perform IVD segmentation based on MR images published up to April 10, 2024. The Quality Assessment of Diagnostic Accuracy Studies-2 tool was used to assess methodological quality, and the pooled dice similarity coefficient (DSC) score and Intersection over Union (IoU) were calculated to evaluate segmentation performance.

RESULTS: 45 studies were included in this systematic review, of which 16 provided complete segmentation performance data and were included in the quantitative meta-analysis. The results indicated that DL models showed satisfactory IVD segmentation performance, with a pooled DSC of 0.900 (95% confidence interval [CI]: 0.887-0.914) and IoU of 0.863 (95% CI: 0.730-0.995). However, the subgroup analysis did not show significant effects of factors on IVD segmentation performance, including network dimensionality, algorithm type, publication year, number of patients, scanning direction, data augmentation, and cross-validation.

CONCLUSIONS: This study highlights the potential of DL technology in IVD segmentation and its further applications. However, due to the heterogeneity in algorithm frameworks and result reporting of the included studies, the conclusions should be interpreted with caution. Future research should focus on training generalized models on large-scale datasets to enhance their clinical application.

PMID:39169382 | DOI:10.1186/s13018-024-05002-5

Brief Bioinform. 2024 Jul 25;25(5):bbae410. doi: 10.1093/bib/bbae410.

ABSTRACT

Immunotherapy with immune checkpoint inhibitors (ICIs) is increasingly used to treat various tumor types. Determining patient responses to ICIs presents a significant clinical challenge. Although components of the tumor microenvironment (TME) are used to predict patient outcomes, comprehensive assessments of the TME are frequently overlooked. Using a top-down approach, the TME was divided into five layers-outcome, immune role, cell, cellular component, and gene. Using this structure, a neural network called TME-NET was developed to predict responses to ICIs. Model parameter weights and cell ablation studies were used to investigate the influence of TME components. The model was developed and evaluated using a pan-cancer cohort of 948 patients across four cancer types, with Area Under the Curve (AUC) and accuracy as performance metrics. Results show that TME-NET surpasses established models such as support vector machine and k-nearest neighbors in AUC and accuracy. Visualization of model parameter weights showed that at the cellular layer, Th1 cells enhance immune responses, whereas myeloid-derived suppressor cells and M2 macrophages show strong immunosuppressive effects. Cell ablation studies further confirmed the impact of these cells. At the gene layer, the transcription factors STAT4 in Th1 cells and IRF4 in M2 macrophages significantly affect TME dynamics. Additionally, the cytokine-encoding genes IFNG from Th1 cells and ARG1 from M2 macrophages are crucial for modulating immune responses within the TME. Survival data from immunotherapy cohorts confirmed the prognostic ability of these markers, with p-values <0.01. In summary, TME-NET performs well in predicting immunotherapy responses and offers interpretable insights into the immunotherapy process. It can be customized at https://immbal.shinyapps.io/TME-NET.

PMID:39167797 | DOI:10.1093/bib/bbae410

JMIR Ment Health. 2024 Aug 21;11:e53714. doi: 10.2196/53714.

ABSTRACT

BACKGROUND: Mental stress and its consequent mental health disorders (MDs) constitute a significant public health issue. With the advent of machine learning (ML), there is potential to harness computational techniques for better understanding and addressing mental stress and MDs. This comprehensive review seeks to elucidate the current ML methodologies used in this domain to pave the way for enhanced detection, prediction, and analysis of mental stress and its subsequent MDs.

OBJECTIVE: This review aims to investigate the scope of ML methodologies used in the detection, prediction, and analysis of mental stress and its consequent MDs.

METHODS: Using a rigorous scoping review process with PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines, this investigation delves into the latest ML algorithms, preprocessing techniques, and data types used in the context of stress and stress-related MDs.

RESULTS: A total of 98 peer-reviewed publications were examined for this review. The findings highlight that support vector machine, neural network, and random forest models consistently exhibited superior accuracy and robustness among all ML algorithms examined. Physiological parameters such as heart rate measurements and skin response are prevalently used as stress predictors due to their rich explanatory information concerning stress and stress-related MDs, as well as the relative ease of data acquisition. The application of dimensionality reduction techniques, including mappings, feature selection, filtering, and noise reduction, is frequently observed as a crucial step preceding the training of ML algorithms.

CONCLUSIONS: The synthesis of this review identified significant research gaps and outlines future directions for the field. These encompass areas such as model interpretability, model personalization, the incorporation of naturalistic settings, and real-time processing capabilities for the detection and prediction of stress and stress-related MDs.

PMID:39167782 | DOI:10.2196/53714

IEEE Trans Med Imaging. 2024 Aug 21;PP. doi: 10.1109/TMI.2024.3447214. Online ahead of print.

ABSTRACT

Histological image registration is a fundamental task in histological image analysis. It is challenging because of substantial appearance differences due to multiple staining. Keypoint correspondences, i.e., matched keypoint pairs, have been introduced to guide unsupervised deep learning (DL) based registration methods to handle such a registration task. This paper proposes an iterative keypoint correspondence-guided (IKCG) unsupervised network for non-rigid histological image registration. Fixed deep features and learnable deep features are introduced as keypoint descriptors to automatically establish keypoint correspondences, the distance between which is used as a loss function to train the registration network. Fixed deep features extracted from DL networks that are pre-trained on natural image datasets are more discriminative than handcrafted ones, benefiting from the deep and hierarchical nature of DL networks. The intermediate layer outputs of the registration networks trained on histological image datasets are extracted as learnable deep features, which reveal unique information for histological images. An iterative training strategy is adopted to train the registration network and optimize learnable deep features jointly. Benefiting from the excellent matching ability of learnable deep features optimized with the iterative training strategy, the proposed method can solve the local non-rigid large displacement problem, an inevitable problem usually caused by misoperation, such as tears in producing tissue slices. The proposed method is evaluated on the Automatic Non-rigid Histology Image Registration (ANHIR) website and AutomatiC Registration Of Breast cAncer Tissue (ACROBAT) website. It ranked 1st on both websites as of August 6th, 2024.

PMID:39167523 | DOI:10.1109/TMI.2024.3447214

IEEE/ACM Trans Comput Biol Bioinform. 2024 Aug 21;PP. doi: 10.1109/TCBB.2024.3447110. Online ahead of print.

ABSTRACT

CircRNA is closely related to human disease, so it is important to predict circRNA-disease association (CDA). However, the traditional biological detection methods have high difficulty and low accuracy, and computational methods represented by deep learning ignore the ability of the model to explicitly extract local depth information of the CDA. We propose a model based on knowledge graph from recursion and attention aggregation for circRNA-disease association prediction (KGRACDA). This model combines explicit structural features and implicit embedding information of graphs, optimizing graph embedding vectors. First, we built large-scale, multi-source heterogeneous datasets and construct a knowledge graph of multiple RNAs and diseases. After that, we use a recursive method to build multi-hop subgraphs and optimize graph attention mechanism by gating mechanism, mining local depth information. At the same time, the model uses multi-head attention mechanism to balance global and local depth features of graphs, and generate CDA prediction scores. KGRACDA surpasses other methods by capturing local and global depth information related to CDA. We update an interactive web platform HNRBase v2.0, which visualizes circRNA data, and allows users to download data and predict CDA using model.

PMID:39167510 | DOI:10.1109/TCBB.2024.3447110

IEEE/ACM Trans Comput Biol Bioinform. 2024 Aug 21;PP. doi: 10.1109/TCBB.2024.3447037. Online ahead of print.

ABSTRACT

The function labeling of enzymes has a wide range of application value in the medical field, industrial biology and other fields. Scientists define enzyme categories by enzyme commission (EC) numbers. At present, although there are some tools for enzyme function prediction, their effects have not reached the application level. To improve the precision of enzyme function prediction, we propose a parallel convolutional contrastive learning (PCCL) method to predict enzyme functions. First, we use the advanced protein language model ESM-2 to preprocess the protein sequences. Second, PCCL combines convolutional neural networks (CNNs) and contrastive learning to improve the prediction precision of multifunctional enzymes. Contrastive learning can make the model better deal with the problem of class imbalance. Finally, the deep learning framework is mainly composed of three parallel CNNs for fully extracting sample features. we compare PCCL with state-of-art enzyme function prediction methods based on three evaluation metrics. The performance of our model improves on both two test sets. Especially on the smaller test set, PCCL improves the AUC by 2.57%. The source code can be downloaded from https://github.com/biomg/PCCL.

PMID:39167509 | DOI:10.1109/TCBB.2024.3447037

IEEE Trans Pattern Anal Mach Intell. 2024 Aug 21;PP. doi: 10.1109/TPAMI.2024.3446949. Online ahead of print.

ABSTRACT

Continual learning, also known as incremental learning or life-long learning, stands at the forefront of deep learning and AI systems. It breaks through the obstacle of one-way training on close sets and enables continuous adaptive learning on open-set conditions. In the recent decade, continual learning has been explored and applied in multiple fields especially in computer vision covering classification, detection and segmentation tasks. Continual semantic segmentation (CSS), of which the dense prediction peculiarity makes it a challenging, intricate and burgeoning task. In this paper, we present a review of CSS, committing to building a comprehensive survey on problem formulations, primary challenges, universal datasets, neoteric theories and multifarious applications. Concretely, we begin by elucidating the problem definitions and primary challenges. Based on an in-depth investigation of relevant approaches, we sort out and categorize current CSS models into two main branches including data-replay and data-free sets. In each branch, the corresponding approaches are similarity-based clustered and thoroughly analyzed, following qualitative comparison and quantitative reproductions on relevant datasets. Besides, we also introduce four CSS specialities with diverse application scenarios and development tendencies. Furthermore, we develop a benchmark for CSS encompassing representative references, evaluation results and reproductions, which is available at https://github.com/YBIO/SurveyCSS. We hope this survey can serve as a reference-worthy and stimulating contribution to the advancement of the life-long learning field, while also providing valuable perspectives for related fields.

PMID:39167503 | DOI:10.1109/TPAMI.2024.3446949

Kidney360. 2024 Aug 21. doi: 10.34067/KID.0000000000000556. Online ahead of print.

ABSTRACT

Traditional medical Artificial Intelligence models, approved for clinical use, restrict themselves to single-modal data e.g. images only, limiting their applicability in the complex, multimodal environment of medical diagnosis and treatment. Multimodal Transformer Models in healthcare can effectively process and interpret diverse data forms such as text, images, and structured data. They have demonstrated impressive performance on standard benchmarks like USLME question banks and continue to improve with scale. However, the adoption of these advanced AI models is not without challenges. While multimodal deep learning models like Transformers offer promising advancements in healthcare, their integration requires careful consideration of the accompanying ethical and environmental challenges.

PMID:39167446 | DOI:10.34067/KID.0000000000000556

J Med Internet Res. 2024 Aug 21;26:e52730. doi: 10.2196/52730.

ABSTRACT

BACKGROUND: Accurate patient outcome prediction in the intensive care unit (ICU) can potentially lead to more effective and efficient patient care. Deep learning models are capable of learning from data to accurately predict patient outcomes, but they typically require large amounts of data and computational resources. Transfer learning (TL) can help in scenarios where data and computational resources are scarce by leveraging pretrained models. While TL has been widely used in medical imaging and natural language processing, it has been rare in electronic health record (EHR) analysis. Furthermore, domain adaptation (DA) has been the most common TL method in general, whereas inductive transfer learning (ITL) has been rare. To the best of our knowledge, DA and ITL have never been studied in-depth in the context of EHR-based ICU patient outcome prediction.

OBJECTIVE: This study investigated DA, as well as rarely researched ITL, in EHR-based ICU patient outcome prediction under simulated, varying levels of data scarcity.

METHODS: Two patient cohorts were used in this study: (1) eCritical, a multicenter ICU data from 55,689 unique admission records from 48,672 unique patients admitted to 15 medical-surgical ICUs in Alberta, Canada, between March 2013 and December 2019, and (2) Medical Information Mart for Intensive Care III, a single-center, publicly available ICU data set from Boston, Massachusetts, acquired between 2001 and 2012 containing 61,532 admission records from 46,476 patients. We compared DA and ITL models with baseline models (without TL) of fully connected neural networks, logistic regression, and lasso regression in the prediction of 30-day mortality, acute kidney injury, ICU length of stay, and hospital length of stay. Random subsets of training data, ranging from 1% to 75%, as well as the full data set, were used to compare the performances of DA and ITL with the baseline models at various levels of data scarcity.

RESULTS: Overall, the ITL models outperformed the baseline models in 55 of 56 comparisons (all P values <.001). The DA models outperformed the baseline models in 45 of 56 comparisons (all P values <.001). ITL resulted in better performance than DA in terms of the number of times and the margin with which it outperformed the baseline models. In 11 of 16 cases (8 of 8 for ITL and 3 of 8 for DA), TL models outperformed baseline models when trained using 1% data subset.

CONCLUSIONS: TL-based ICU patient outcome prediction models are useful in data-scarce scenarios. The results of this study can be used to estimate ICU outcome prediction performance at different levels of data scarcity, with and without TL. The publicly available pretrained models from this study can serve as building blocks in further research for the development and validation of models in other ICU cohorts and outcomes.

PMID:39167442 | DOI:10.2196/52730

JMIR Form Res. 2024 Aug 21;8:e55641. doi: 10.2196/55641.

ABSTRACT

BACKGROUND: Artificial intelligence (AI) based computer-aided detection devices are recommended for screening and triaging of pulmonary tuberculosis (TB) using digital chest x-ray (CXR) images (soft copies). Most AI algorithms are trained using input data from digital CXR Digital Imaging and Communications in Medicine (DICOM) files. There can be scenarios when only digital CXR films (hard copies) are available for interpretation. A smartphone-captured photo of the digital CXR film may be used for AI to process in such a scenario. There is a gap in the literature investigating if there is a significant difference in the performance of AI algorithms when digital CXR DICOM files are used as input for AI to process as opposed to photos of the digital CXR films being used as input.

OBJECTIVE: The primary objective was to compare the agreement of AI in detecting radiological signs of TB when using DICOM files (denoted as CXRd) as input versus when using smartphone-captured photos of digital CXR films (denoted as CXRp) with human readers.

METHODS: Pairs of CXRd and CXRp images were obtained retrospectively from patients screened for TB. AI results were obtained using both the CXRd and CXRp files. The majority consensus on the presence or absence of TB in CXR pairs was obtained from a panel of 3 independent radiologists. The positive and negative percent agreement of AI in detecting radiological signs of TB in CXRd and CXRp were estimated by comparing with the majority consensus. The distribution of AI probability scores was also compared.

RESULTS: A total of 1278 CXR pairs were analyzed. The positive percent agreement of AI was found to be 92.22% (95% CI 89.94-94.12) and 90.75% (95% CI 88.32-92.82), respectively, for CXRd and CXRp images (P=.09). The negative percent agreement of AI was 82.08% (95% CI 78.76-85.07) and 79.23% (95% CI 75.75-82.42), respectively, for CXRd and CXRp images (P=.06). The median of the AI probability score was 0.72 (IQR 0.11-0.97) in CXRd and 0.72 (IQR 0.14-0.96) in CXRp images (P=.75).

CONCLUSIONS: We did not observe any statistically significant differences in the output of AI in digital CXRs and photos of digital CXR films.

PMID:39167435 | DOI:10.2196/55641

J Proteome Res. 2024 Aug 21. doi: 10.1021/acs.jproteome.4c00118. Online ahead of print.

ABSTRACT

Proteome analysis currently heavily relies on tandem mass spectrometry (MS/MS), which does not fully utilize MS1 features, as many precursors remain unselected for MS/MS fragmentation, especially in the cases of low abundance samples and wide abundance dynamic range samples. Therefore, leveraging MS1 features as a complement to MS/MS has become an attractive option to improve the coverage of feature identification. Herein, we propose MonoMS1, an approach combining deep learning-based retention time, ion mobility, detectability prediction, and logistic regression-based scoring for MS1 feature identification. The approach achieved a significant increase in MS1 feature identification based on an E. coli data set. Application of MonoMS1 to data sets with wide dynamic range, such as human serum proteome samples, and with low sample abundance, such as single-cell proteome samples, enabled substantial complementation of MS/MS-based peptide and protein identification. This method opens a new avenue for proteomic analysis and can boost proteomic research on complex samples.

PMID:39167422 | DOI:10.1021/acs.jproteome.4c00118

Fa Yi Xue Za Zhi. 2024 Jun 25;40(3):245-253. doi: 10.12116/j.issn.1004-5619.2023.430803.

ABSTRACT

OBJECTIVES: To describe the current state of research and future research hotspots through a metrological analysis of the literature in the field of forensic anthropological remains identification research.

METHODS: The data retrieved and extracted from the Web of Science Core Collection (WoSCC), the core database of the Web of Science information service platform (hereinafter referred to as "WoS"), was used to analyze the trends and topic changes in research on forensic identification of human remains from 1991 to 2022. Network visualisation of publication trends, countries (regions), institutions, authors and topics related to the identification of remains in forensic anthropology was analysed using python 3.9.2 and Gephi 0.10.

RESULTS: A total of 873 papers written in English in the field of forensic anthropological remains identification research were obtained. The journal with the largest number of publications was Forensic Science International (164 articles). The country (region) with the largest number of published papers was China (90 articles). Katholieke Univ Leuven (Netherlands, 21 articles) was the institution with the largest number of publications. Topic analysis revealed that the focus of forensic anthropological remains identification research was sex estimation and age estimation, and the most commonly studied remains were teeth.

CONCLUSIONS: The volume of publications in the field of forensic anthropological remains identification research has a distinct phasing. However, the scope of both international and domestic collaborations remains limited. Traditionally, human remains identification has primarily relied on key areas such as the pelvis, skull, and teeth. Looking ahead, future research will likely focus on the more accurate and efficient identification of multiple skeletal remains through the use of machine learning and deep learning techniques.

PMID:39166305 | DOI:10.12116/j.issn.1004-5619.2023.430803