Artificial Intelligence-Based Methods for Precision Medicine: Diabetes Risk Prediction
Authors:
Farida Mohsen,
Hamada R. H. Al-Absi,
Noha A. Yousri,
Nady El Hajj,
Zubair Shah
Abstract:
The rising prevalence of type 2 diabetes mellitus (T2DM) necessitates the development of predictive models for T2DM risk assessment. Artificial intelligence (AI) models are being extensively used for this purpose, but a comprehensive review of their advancements and challenges is lacking. This sco** review analyzes existing literature on AI-based models for T2DM risk prediction. Forty studies we…
▽ More
The rising prevalence of type 2 diabetes mellitus (T2DM) necessitates the development of predictive models for T2DM risk assessment. Artificial intelligence (AI) models are being extensively used for this purpose, but a comprehensive review of their advancements and challenges is lacking. This sco** review analyzes existing literature on AI-based models for T2DM risk prediction. Forty studies were included, mainly published in the past four years. Traditional machine learning models were more prevalent than deep learning models. Electronic health records were the most commonly used data source. Unimodal AI models relying on EHR data were prominent, while only a few utilized multimodal models. Both unimodal and multimodal models showed promising performance, with the latter outperforming the former. Internal validation was common, while external validation was limited. Interpretability methods were reported in half of the studies. Few studies reported novel biomarkers, and open-source code availability was limited. This review provides insights into the current state and limitations of AI-based T2DM risk prediction models and highlights challenges for their development and clinical implementation.
△ Less
Submitted 24 May, 2023;
originally announced May 2023.
Artificial Intelligence-Based Methods for Fusion of Electronic Health Records and Imaging Data
Authors:
Farida Mohsen,
Hazrat Ali,
Nady El Hajj,
Zubair Shah
Abstract:
Healthcare data are inherently multimodal, including electronic health records (EHR), medical images, and multi-omics data. Combining these multimodal data sources contributes to a better understanding of human health and provides optimal personalized healthcare. Advances in artificial intelligence (AI) technologies, particularly machine learning (ML), enable the fusion of these different data mod…
▽ More
Healthcare data are inherently multimodal, including electronic health records (EHR), medical images, and multi-omics data. Combining these multimodal data sources contributes to a better understanding of human health and provides optimal personalized healthcare. Advances in artificial intelligence (AI) technologies, particularly machine learning (ML), enable the fusion of these different data modalities to provide multimodal insights. To this end, in this sco** review, we focus on synthesizing and analyzing the literature that uses AI techniques to fuse multimodal medical data for different clinical applications. More specifically, we focus on studies that only fused EHR with medical imaging data to develop various AI methods for clinical applications. We present a comprehensive analysis of the various fusion strategies, the diseases and clinical outcomes for which multimodal fusion was used, the ML algorithms used to perform multimodal fusion for each clinical application, and the available multimodal medical datasets. We followed the PRISMA-ScR guidelines. We searched Embase, PubMed, Scopus, and Google Scholar to retrieve relevant studies. We extracted data from 34 studies that fulfilled the inclusion criteria. In our analysis, a typical workflow was observed: feeding raw data, fusing different data modalities by applying conventional machine learning (ML) or deep learning (DL) algorithms, and finally, evaluating the multimodal fusion through clinical outcome predictions. Specifically, early fusion was the most used technique in most applications for multimodal learning (22 out of 34 studies). We found that multimodality fusion models outperformed traditional single-modality models for the same task. Disease diagnosis and prediction were the most common clinical outcomes (reported in 20 and 10 studies, respectively) from a clinical outcome perspective.
△ Less
Submitted 23 October, 2022;
originally announced October 2022.
