NPJ Digital Medicine最新文献

Differentiating between diabetic nephropathy (DN) and non-diabetic renal disease (NDRD) without a kidney biopsy remains a major challenge, often leading to missed opportunities for targeted treatments that could greatly improve NDRD outcomes. To reform the traditional biopsy-all diagnostic paradigm and avoid unnecessary biopsy, we developed a transformer-based deep learning (DL) system for detecting DN and NDRD upon non-invasive multi-modal data of fundus images and clinical characteristics. Our Trans-MUF achieved an AUC of 0.980 (95% CI: 0.979 to 0.980) over the internal retrospective set and also had superior generalizability over a prospective dataset (AUC: 0.989, 95% CI: 0.987 to 0.990) and a multicenter, cross-machine and multi-operator dataset (AUC: 0.932, 95% CI: 0.931 to 0.939). Moreover, the nephrologists‘ diagnosis accuracy can be improved by 21%, through visualization assistance of the DL system. This paper lays a foundation for automatically differentiating DN and NDRD without biopsy. (Registry name: Correlation Study Between Clinical Phenotype and Pathology of Type 2 Diabetic Nephropathy. ID: NCT03865914. Date: 2017-11-30).

Particulate matter (PM) exposure can reduce heart rate variability (HRV), a cardiovascular health marker. This study examines PM_1.0 (aerodynamic diameters <1 μm), PM_2.5 (≥1 μm and <2.5 μm), and PM₁₀ (≥2.5 μm and <10 μm) effects on HRV in patients with environmental diseases as chronic disease groups and vulnerable populations as control groups. PM levels were measured indoors and outdoors for five days in 97 participants, with 24-h HRV monitoring via wearable devices. PM exposure was assessed by categorizing daily cumulative PM concentrations into higher and lower exposure days, while daily average PM concentrations were used for analysis. Results showed significant negative associations between exposure to single and mixtures of different PM metrics and HRV across all groups, particularly in chronic airway disease and higher air pollution exposed groups. These findings highlight that even lower PM levels may reduce HRV, suggesting a need for stricter standards to protect sensitive individuals.

We propose the addition of usability validation to the extended V3 framework, now “V3+”, and describe a pragmatic approach to ensuring that sensor-based digital health technologies can be used optimally at scale by diverse users. Alongside the original V3 components (verification; analytical validation; clinical validation), usability validation will ensure user-centricity of digital measurement tools, paving the way for more inclusive, reliable, and trustworthy digital measures within clinical research and clinical care.

This study developed and evaluated a model for assessing pain during the surgical period using photoplethysmogram data from 242 patients. Pain levels were measured at 2 min intervals using a numerical rating scale or clinical criteria: preoperative, before and after intubation, before and after skin incision, and postoperative. Key features from the photoplethysmography waveform were extracted to build XGBoost-based models for intraoperative and postoperative pain assessment. The combined perioperative model was compared with a commercial surgical pain index, yielding area under the receiver operating characteristics curve scores of 0.819 and 0.927 for intraoperative and postoperative periods, respectively, compared to the commercial index’s scores of 0.829 and 0.577. These results highlight the models’ effectiveness in pain assessment throughout the surgical process, identifying waveform skewness and diastolic phase rate decrease as critical for intraoperative pain assessment and systolic phase area or baseline fluctuation as significant for postoperative pain assessment.

Clinical trial registration: Registration name: Clinical Research Information Service (CRIS). Registration site: http://cris.nih.go.kr. Number: KCT0005840. Principal Investigator: Dr. Byung-Moon Choi. Date of registration: January 28, 2021

Generating synthetic data from medical records is a complex task intensified by patient privacy concerns. In recent years, multiple approaches have been reported for the generation of synthetic data, however, limited attention was given to jointly evaluate the quality and the privacy of the generated data. The quality and privacy of synthetic data stem from multivariate associations across variables, which cannot be assessed by comparing univariate distributions with the original data. Here, we introduce a novel algorithm (MIIC-SDG) for generating synthetic data from electronic records based on a multivariate information framework and Bayesian network theory. We also propose a new metric to quantitatively assess the trade-off between the Quality and Privacy Scores (QPS) of synthetic data generation methods. The performance of MIIC-SDG is demonstrated on different clinical datasets and favorably compares with state-of-the-art synthetic data generation methods, based on the QPS trade-off between several quality and privacy metrics.

This randomized, controlled trial evaluated the impact of plaque visualization combined with daily tasks on cardiovascular risk profile and included 240 participants with coronary arterial disease. The intervention group received the PreventiPlaque app during the 12-month study period in addition to standard care. The app contained daily tasks that promoted lifestyle modifications and adherence to prescribed medication. It included ultrasound images of participants´ individual carotid plaque, which were updated regularly. The impact of plaque visualization and personalized app usage was evaluated, using a change in the SCORE2 as a primary endpoint. In the intervention group, the SCORE2 was significantly lower after the study period (t(120) = 6.43, p_adj < 0.001, d_RM = 0.58). This demonstrates the efficacy of the PreventiPlaque app in supporting lifestyle modifications and medication adherence. These findings suggest that personalized mHealth interventions in combination with visual risk communication are valuable tools in secondary prevention. Trial Registration: The study was registered at ClinicalTrials.gov under the identifier NCT05096637 on 27 October 2021 and was approved by the local ethics committee of the University of Duisburg-Essen (20-9157-BO).

The Digital Personal Data Protection Act (DPDPA), 2023 of India provides a regulatory framework for use and security of personal digital data. However, instances, wherein the patients consult the clinicians via digital means of communication: the implications of DPDPA, 2023 for the medical personnel remain unclear. This paper critically discusses the gray areas encountered in the Indian medical ecosystem and DPDPA, 2023 and; lists the recommendations to address them.

Glaucoma is the leading cause of irreversible blindness globally. Research indicates a disproportionate impact of glaucoma on racial and ethnic minorities. Existing deep learning models for glaucoma detection might not achieve equitable performance across diverse identity groups. We developed fair identify normalization (FIN) module to equalize the feature importance across different identity groups to improve model performance equity. The optical coherence tomography (OCT) measurements were used to categorize patients into glaucoma and non-glaucoma. The equity-scaled area under the receiver operating characteristic curve (ES-AUC) was adopted to quantify model performance equity. With FIN for racial groups, the overall AUC and ES-AUC increased from 0.82 to 0.85 and 0.77 to 0.81, respectively, with the AUC for Blacks increasing from 0.77 to 0.82. With FIN for ethnic groups, the overall AUC and ES-AUC increased from 0.82 to 0.84 and 0.77 to 0.80, respectively, with the AUC for Hispanics increasing from 0.75 to 0.79.

While drug combinations are increasingly important in disease treatment, predicting their therapeutic interactions remains challenging. This review systematically analyzes computational methods for predicting drug combination effects through multi-omics data integration. We comprehensively assess key algorithms including DrugComboRanker and AuDNNsynergy, and evaluate integration approaches encompassing kernel regression and graph networks. The review elucidates artificial intelligence applications in predicting drug synergistic and antagonistic effects.