JMIR Publications

ABSTRACT

Background:

Oral microenvironmental disorders are associated with an increased risk of heart failure with preserved ejection fraction (HFpEF). The technology of hyperspectral imaging (HSI) enables the detection of substances that are visually indistinguishable to the human eye, providing a non-invasive approach with extensive applications in medical diagnostics.

Objective:

The objective of this study is to develop and validate a digital, non-invasive oral diagnostic model for patients with HFpEF using hyperspectral imaging combined with various machine learning algorithms.

Methods:

Between April 2023 and August 2023, a total of 140 patients were recruited from Renmin Hospital of Wuhan University to serve as the training and internal testing groups for this study. Subsequently, from August 2024 to September 2024, an additional 35 patients were enrolled from Three Gorges University and Yichang Central People's Hospital to constitute the external testing group. After preprocessing to ensure image quality, spectral and textural features were extracted from the images. We extracted 25 spectral bands from each patient image and obtained 8 corresponding texture features for evaluating the performance of machine learning algorithms. 28 machine learning algorithms were evaluated for their ability to distinguish control subjects and HFpEF patients. The model demonstrating optimal performance in both internal and external testing was selected for constructing the HFpEF diagnostic model. Hyperspectral bands significant for differentiating HFpEF patients were identified for further interpretative analysis. The Shapley Additive exPlanations (SHAP) model was employed to provide analytical insights into feature importance.

Results:

Participants were divided into a training group (n=105), internal testing group (n=35), and external testing group (n=35), with consistent baseline characteristics across groups. Among the 28 algorithms tested, the Random Forest algorithm demonstrated superior performance with an area under the receiver operating characteristic curve (AUC) of 0.884 and an accuracy of 82.9% in the internal testing group, as well as an AUC of 0.812 and an accuracy of 85.7% in the external testing group. For model interpretation, we employed the top 25 features identified by the Random Forest algorithm. The application of SHAP analysis revealed discernible distinctions between control subjects and HFpEF patients, thereby validating the diagnostic model's capacity to accurately discriminate HFpEF patients.

Conclusions:

This non-invasive and efficient model facilitates the identification of patients with HFpEF, thereby promoting early detection, diagnosis, and treatment. Our research presents a clinically advanced diagnostic framework for HFpEF, validated using independent datasets, which demonstrates significant potential to enhance patient care. Clinical Trial: The study was registered on the China Clinical Trial Registry (No.ChiCTR2300078855).