UNASSIGNED: Osteoporosis, characterized by low bone mineral density (BMD), is an increasingly serious public health issue. So far, several traditional regression models and machine learning (ML) algorithms have been proposed for predicting osteoporosis risk. However, these models have shown relatively low accuracy in clinical implementation. Recently proposed deep learning (DL) approaches, such as deep neural network (DNN), which can discover knowledge from complex hidden interactions, offer a new opportunity to improve predictive performance. In this study, we aimed to assess whether DNN can achieve a better performance in osteoporosis risk prediction.
UNASSIGNED: By utilizing hip BMD and extensive demographic and routine clinical data of 8,134 subjects with age more than 40 from the Louisiana Osteoporosis Study (LOS), we developed and constructed a novel DNN framework for predicting osteoporosis risk and compared its performance in osteoporosis risk prediction with four conventional ML models, namely random forest (RF), artificial neural network (ANN), k-nearest neighbor (KNN), and support vector machine (SVM), as well as a traditional regression model termed osteoporosis self-assessment tool (OST). Model performance was assessed by area under \'receiver operating curve\' (AUC) and accuracy.
UNASSIGNED: By using 16 discriminative variables, we observed that the DNN approach achieved the best predictive performance (AUC = 0.848) in classifying osteoporosis (hip BMD T-score ≤ -1.0) and non-osteoporosis risk (hip BMD T-score > -1.0) subjects, compared to the other approaches. Feature importance analysis showed that the top 10 most important variables identified by the DNN model were weight, age, gender, grip strength, height, beer drinking, diastolic pressure, alcohol drinking, smoke years, and economic level. Furthermore, we performed subsampling analysis to assess the effects of varying number of sample size and variables on the predictive performance of these tested models. Notably, we observed that the DNN model performed equally well (AUC = 0.846) even by utilizing only the top 10 most important variables for osteoporosis risk prediction. Meanwhile, the DNN model can still achieve a high predictive performance (AUC = 0.826) when sample size was reduced to 50% of the original dataset.
UNASSIGNED: In conclusion, we developed a novel DNN model which was considered to be an effective algorithm for early diagnosis and intervention of osteoporosis in the aging population.

OBJECTIVE: Develop a novel technique to identify an optimal number of regression units corresponding to a single risk point, while creating risk scoring systems from logistic regression-based disease predictive models. The optimal value of this hyperparameter balances simplicity and accuracy, yielding risk scores of small scale and high accuracy for patient risk stratification.
METHODS: The proposed technique applies an adapted line search across all potential hyperparameter values. Additionally, DeLong test is integrated to ensure the selected value produces an accuracy insignificantly different from the best achievable risk score accuracy. We assessed the approach through two case studies predicting diabetic retinopathy (DR) within six months and hip fracture readmissions (HFR) within 30 days, involving cohorts of 90 400 diabetic patients and 18 065 hip fracture patients.
RESULTS: Our scores achieve accuracies insignificantly different from those obtained by existing approaches, reaching AUROCs of 0.803 and 0.645 for DR and HFR predictions, respectively. Regarding the scale, our scores ranged 0-53 for DR and 0-15 for HFR, while scores produced by existing methods frequently spanned hundreds or thousands.
CONCLUSIONS: According to the assessment, our risk scores offer simple and accurate predictions for diseases. Furthermore, our new DR score provides a competitive alternative to state-of-the-art risk scores for DR, while our HFR case study presents the first risk score for this condition.
CONCLUSIONS: Our technique offers a generalizable framework for crafting precise risk scores of compact scales, addressing the demand for user-friendly and effective risk stratification tool in healthcare.

Graph neural networks (GNNs) have gained significant attention in disease prediction where the latent embeddings of patients are modeled as nodes and the similarities among patients are represented through edges. The graph structure, which determines how information is aggregated and propagated, plays a crucial role in graph learning. Recent approaches typically create graphs based on patients\' latent embeddings, which may not accurately reflect their real-world closeness. Our analysis reveals that raw data, such as demographic attributes and laboratory results, offers a wealth of information for assessing patient similarities and can serve as a compensatory measure for graphs constructed exclusively from latent embeddings. In this study, we first construct adaptive graphs from both latent representations and raw data respectively, and then merge these graphs via weighted summation. Given that the graphs may contain extraneous and noisy connections, we apply degree-sensitive edge pruning and kNN sparsification techniques to selectively sparsify and prune these edges. We conducted intensive experiments on two diagnostic prediction datasets, and the results demonstrate that our proposed method surpasses current state-of-the-art techniques.

BACKGROUND: Chronic disease management is a major health issue worldwide. With the paradigm shift to preventive medicine, disease prediction modeling using machine learning is gaining importance for precise and accurate medical judgement.
OBJECTIVE: This study aimed to develop high-performance prediction models for 4 chronic diseases using the common data model (CDM) and machine learning and to confirm the possibility for the extension of the proposed models.
METHODS: In this study, 4 major chronic diseases-namely, diabetes, hypertension, hyperlipidemia, and cardiovascular disease-were selected, and a model for predicting their occurrence within 10 years was developed. For model development, the Atlas analysis tool was used to define the chronic disease to be predicted, and data were extracted from the CDM according to the defined conditions. A model for predicting each disease was built with 4 algorithms verified in previous studies, and the performance was compared after applying a grid search.
RESULTS: For the prediction of each disease, we applied 4 algorithms (logistic regression, gradient boosting, random forest, and extreme gradient boosting), and all models show greater than 80% accuracy. As compared to the optimized model\'s performance, extreme gradient boosting presented the highest predictive performance for the 4 diseases (diabetes, hypertension, hyperlipidemia, and cardiovascular disease) with 80% or greater and from 0.84 to 0.93 in area under the curve standards.
CONCLUSIONS: This study demonstrates the possibility for the preemptive management of chronic diseases by predicting the occurrence of chronic diseases using the CDM and machine learning. With these models, the risk of developing major chronic diseases within 10 years can be demonstrated by identifying health risk factors using our chronic disease prediction machine learning model developed with the real-world data-based CDM and National Health Insurance Corporation examination data that individuals can easily obtain.

Data acquisition for transcriptomic studies used to be the bottleneck in the transcriptomic analytical pipeline. However, recent developments in transcriptome profiling technologies have increased researchers\' ability to obtain data, resulting in a shift in focus to data analysis. Incorporating machine learning to traditional analytical methods allows the possibility of handling larger volumes of complex data more efficiently. Many bioinformaticians, especially those unfamiliar with ML in the study of human transcriptomics and complex biological systems, face a significant barrier stemming from their limited awareness of the current landscape of ML utilisation in this field. To address this gap, this review endeavours to introduce those individuals to the general types of ML, followed by a comprehensive range of more specific techniques, demonstrated through examples of their incorporation into analytical pipelines for human transcriptome investigations. Important computational aspects such as data pre-processing, task formulation, results (performance of ML models), and validation methods are encompassed. In hope of better practical relevance, there is a strong focus on studies published within the last five years, almost exclusively examining human transcriptomes, with outcomes compared with standard non-ML tools.

OBJECTIVE: Liver disease causes two million deaths annually, accounting for 4% of all deaths globally. Prediction or early detection of the disease via machine learning algorithms on large clinical data have become promising and potentially powerful, but such methods often have some limitations due to the complexity of the data. In this regard, ensemble learning has shown promising results. There is an urgent need to evaluate different algorithms and then suggest a robust ensemble algorithm in liver disease prediction.
METHODS: Three ensemble approaches with nine algorithms are evaluated on a large dataset of liver patients comprising 30,691 samples with 11 features. Various preprocessing procedures are utilized to feed the proposed model with better quality data, in addition to the appropriate tuning of hyperparameters and selection of features.
RESULTS: The models\' performances with each algorithm are extensively evaluated with several positive and negative performance metrics along with runtime. Gradient boosting is found to have the overall best performance with 98.80% accuracy and 98.50% precision, recall and F1-score for each.
CONCLUSIONS: The proposed model with gradient boosting bettered in most metrics compared with several recent similar works, suggesting its efficacy in predicting liver disease. It can be further applied to predict other diseases with the commonality of predicate indicators.

BACKGROUND: Accurate prediction of an individual\'s predisposition to diseases is vital for preventive medicine and early intervention. Various statistical and machine learning models have been developed for disease prediction using clinico-genomic data. However, the accuracy of clinico-genomic prediction of diseases may vary significantly across ancestry groups due to their unequal representation in clinical genomic datasets.
METHODS: We introduced a deep transfer learning approach to improve the performance of clinico-genomic prediction models for data-disadvantaged ancestry groups. We conducted machine learning experiments on multi-ancestral genomic datasets of lung cancer, prostate cancer, and Alzheimer\'s disease, as well as on synthetic datasets with built-in data inequality and distribution shifts across ancestry groups.
RESULTS: Deep transfer learning significantly improved disease prediction accuracy for data-disadvantaged populations in our multi-ancestral machine learning experiments. In contrast, transfer learning based on linear frameworks did not achieve comparable improvements for these data-disadvantaged populations.
CONCLUSIONS: This study shows that deep transfer learning can enhance fairness in multi-ancestral machine learning by improving prediction accuracy for data-disadvantaged populations without compromising prediction accuracy for other populations, thus providing a Pareto improvement towards equitable clinico-genomic prediction of diseases.

OBJECTIVE: Autoantibodies against hexokinase 1 (HK1) were recently proposed to be associated with diabetic macular edema (DME). We hypothesized that anti-HK1 autoantibodies can be used as DME markers and to predict DME onset.
METHODS: Serum from patients with 1) DME, 2) diabetes mellitus (DM), 3) allergies or autoimmunities, and 4) control subjects was tested for anti-HK1 and anti-hexokinase 2 (HK2) autoantibodies by immunoblotting. Patients with DM were prospectively followed for up to nine years, and the association of anti-HK1 antibodies with new-onset DME was evaluated. The vitreous humor was also tested for autoantibodies.
RESULTS: Among patients with DME, 32 % were positive for anti-HK1 autoantibodies (42 % of those with underlying type 1 DM and 31 % of those with underlying type 2 DM), and 12 % were positive for anti-HK2 autoantibodies, with only partial overlap of these two groups of patients. Anti-HK1 positive were also 7 % of patients with DM, 6 % of patients with allergies and autoimmunities, and 3 % of control subjects. The latter three groups were anti-HK2 negative. Only one of seven patients with DM who were initially anti-HK1 positive developed DME.
CONCLUSIONS: Anti-HK1 autoantibodies can be used as DME markers but fail to predict DME onset.

BACKGROUND: Medical records are a valuable source for understanding patient health conditions. Doctors often use these records to assess health without solely depending on time-consuming and complex examinations. However, these records may not always be directly relevant to a patient\'s current health issue. For instance, information about common colds may not be relevant to a more specific health condition. While experienced doctors can effectively navigate through unnecessary details in medical records, this excess information presents a challenge for machine learning models in predicting diseases electronically. To address this, we have developed \'al-BERT\', a new disease prediction model that leverages the BERT framework. This model is designed to identify crucial information from medical records and use it to predict diseases. \'al-BERT\' operates on the principle that the structure of sentences in diagnostic records is similar to regular linguistic patterns. However, just as stuttering in speech can introduce \'noise\' or irrelevant information, similar issues can arise in written records, complicating model training. To overcome this, \'al-BERT\' incorporates a semi-supervised layer that filters out irrelevant data from patient visitation records. This process aims to refine the data, resulting in more reliable indicators for disease correlations and enhancing the model\'s predictive accuracy and utility in medical diagnostics.
METHODS: To discern noise diseases within patient records, especially those resembling influenza-like illnesses, our approach employs a customized semi-supervised learning algorithm equipped with a focused attention mechanism. This mechanism is specifically calibrated to enhance the model\'s sensitivity to chronic conditions while concurrently distilling salient features from patient records, thereby augmenting the predictive accuracy and utility of the model in clinical settings. We evaluate the performance of al-BERT using real-world health insurance data provided by Taiwan\'s National Health Insurance.
RESULTS: In our study, we evaluated our model against two others: one based on BERT that uses complete disease records, and another variant that includes extra filtering techniques. Our findings show that models incorporating filtering mechanisms typically perform better than those using the entire, unfiltered dataset. Our approach resulted in improved outcomes across several key measures: AUC-ROC (an indicator of a model\'s ability to distinguish between classes), precision (the accuracy of positive predictions), recall (the model\'s ability to find all relevant cases), and overall accuracy. Most notably, our model showed a 15% improvement in recall compared to the current best-performing method in the field of disease prediction.
CONCLUSIONS: The conducted ablation study affirms the advantages of our attention mechanism and underscores the crucial role of the selection module within al-BERT.

Myelin oligodendrocyte glycoprotein-antibody-associated disease (MOGAD) is a demyelinating central nervous system disorder. We aimed to uncover immune pathways altered in MOGAD to predict disease progression. Using nanostring nCounter technology, we analyzed immune gene expression in PBMCs from MOGAD patients and compare it with healthy controls (HCs). We found 35 genes that distinguished MOGAD patients and HCs. We then validated those results in a larger cohort including MS and NMOSD patients. Expressions of HLA-DRA was significantly lower in MOGAD patients. This reduction in HLA-DRA, correlated with a monophasic disease course and greater brain volume, enhancing our ability to predict MOGAD progression.