AI-Driven Genotype Prediction Using EHR and Multimodal Data

Predicting Patient Genotypes Using Electronic Health Records and Multimodal Data Through AI-Based Models

The goal of this clinical study is to explore the potential of using electronic health records (EHR) and multimodal data (such as imaging, lab results, and clinical history) to predict a patient's genotype. The study will evaluate whether predictive models based on this non-genetic data can accurately infer genetic information, which traditionally requires direct genetic testing.

Study Overview

• Status: Recruiting
• Conditions: Genotype
• Intervention / Treatment: Other: AI-Predictng Model

Detailed Description

This multi-center, retrospective clinical study aims to evaluate the use of electronic health records (EHR) and multimodal data (such as clinical lab results, imaging data, and medical history) in predicting a patient's genotype. The primary objective of the study is to develop an AI-based prediction model that can infer genetic information by analyzing available health data, eliminating the need for direct genetic testing.The AI model will be trained to process and integrate large datasets, including EHR, lab results, and imaging data such as X-rays, MRIs, and ultrasounds, in order to predict genotypic information. The study will compare the AI-based predictions to actual genetic testing results to evaluate the accuracy of the model. If successful, this method could provide a non-invasive, cost-effective tool for genotype prediction, which could be used in personalized medicine, early disease diagnosis, and risk stratification.Participants will not undergo any genetic testing as part of the study. Instead, their historical medical data will be analyzed by the AI system to predict genetic information and associated disease risks. The study will assess the model's ability to predict genetic predispositions to various health conditions based on the available health data. By doing so, the study aims to advance the use of AI in clinical decision-making and genetic diagnostics.

• Study Type: Observational
• Enrollment (Estimated): 100000

Contacts and Locations

• Study Contact: Name: Fei Liu, MD; Phone Number: +86 13810512704; Email: liufei_2359@163.com

Study Locations

• China
  • Guangdong
    • Guangzhou, Guangdong, China
      • Recruiting
      • Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University
      • Contact: Yunfang Yu; Phone Number: +86 020-81332199; Email: yuyf9@mail.sysu.edu.cn
    • Guangzhou, Guangdong, China
      • Recruiting
      • Sun Yat-sen University Cancer Hospital
      • Contact: Yuxing Lu; Phone Number: +86 13161233730; Email: yxlu0613@gmail.com
  • Zhejiang
    • Wenzhou, Zhejiang, China
      • Recruiting
      • Second Affiliated Hospital of Wenzhou Medical University
      • Contact: Sian Liu; Phone Number: +86-0577-88002888; Email: liusan@mail3.sysu.edu.cn
    • Wenzhou, Zhejiang, China
      • Completed
      • First Affiliated Hospital of Wenzhou Medical University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

The study population will be selected from multiple healthcare centers that maintain comprehensive electronic health records (EHR) and have access to multimodal clinical data, including lab results, medical imaging (e.g., X-rays, MRIs, CT scans), and medical history. Participants will be individuals with a variety of health conditions for which genotype information is relevant, although no specific genetic characteristics will be used for selection. The focus will be on utilizing the available health data to predict genetic information through the AI model. The study aims to evaluate the accuracy and utility of using non-genetic data, such as EHR and multimodal imaging, for predicting patient genotypes, which may provide an alternative approach to traditional genetic testing methods.

Description

Inclusion Criteria:

1. Participants must have comprehensive electronic health records (EHR), including medical history, lab results, and relevant imaging data (e.g., X-rays, MRIs, CT scans).
2. Participants must have existing genetic testing data available for comparison, if applicable.
3. Participants must be willing to provide consent for the use of their health data in the study.
4. Participants must have no active intervention related to genetic testing or prediction during the study period.
5. Participants should have complete and verifiable health data to allow for accurate prediction by the AI model.

Exclusion Criteria:

1. Participants without available EHR, lab results, or imaging data.
2. Participants with ambiguous, inaccurate, or unverifiable genetic testing results that cannot be used for comparison.
3. Patients with significant discrepancies or missing data that would prevent the AI model from making accurate predictions.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Intervention / Treatment

AI-Based Genotype Prediction Using EHR and Multimodal Data; This cohort consists of patients whose historical health data, including electronic health records (EHR), clinical lab results, and multimodal imaging data (such as X-rays, MRIs, and CT scans), will be analyzed by an AI-based prediction model to predict their genotype. There are no active interventions in this cohort, as the study aims to use non-genetic health data to infer genetic information. Participants will not undergo genetic testing but will provide their health data for analysis by the AI system. The goal of this group is to assess the accuracy of the AI model in predicting genotypes and identifying genetic predispositions to various diseases based on available health data.

Other: AI-Predictng Model; The intervention in this study involves an AI-based predictive model designed to analyze and integrate patient electronic health records (EHR), clinical lab results, and multimodal imaging data (e.g., X-rays, MRIs, CT scans). The AI model is trained to predict a patient's genotype based on these non-genetic data sources. This model uses machine learning algorithms to detect patterns and infer genetic information that would traditionally require direct genetic testing. There are no active treatments or genetic tests involved in this intervention; rather, the AI system serves as a tool to predict genetic information from available clinical data, offering a non-invasive and potentially more accessible alternative to genetic testing.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Area Under the Curve (AUC)	AUC of the ROC curve, used to quantify diagnostic accuracy. No unit (a ratio or percentage, typically expressed as a number between 0 and 1).	1 year
F1 Score	The F1 score is the harmonic mean of precision and sensitivity (recall). It is a good measure of the model's ability to identify both true positives and minimize false positives, especially in cases where the classes are imbalanced (e.g., when the number of healthy cases is much higher than disease cases). The F1 score ranges from 0 to 1, with 1 indicating perfect precision and recall.	1 year

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Sensitivity (True Positive Rate)	Sensitivity measures how well the AI model identifies true positive cases, such as correctly diagnosing pregnant women with complications or identifying neonatal disorders.	1 year
Specificity (True Negative Rate)	Specificity measures the ability of the AI model to correctly identify cases without diseases, ensuring that healthy mothers and infants are correctly identified as negative.	1 year

Collaborators and Investigators

• Sponsor: The Eye Hospital of Wenzhou Medical University

Study record dates

Study Major Dates

• Study Start (Actual): July 1, 2023
• Primary Completion (Estimated): June 1, 2025
• Study Completion (Estimated): June 1, 2025

Study Registration Dates

• First Submitted: January 19, 2025
• First Submitted That Met QC Criteria: January 19, 2025
• First Posted (Actual): January 24, 2025

Study Record Updates

• Last Update Posted (Actual): April 17, 2025
• Last Update Submitted That Met QC Criteria: April 16, 2025
• Last Verified: April 1, 2025

More Information

Terms related to this study

• Keywords: Genotype; AI; Multimodal data; AI-prediction
• Other Study ID Numbers: Genotype

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No