AI Is Effective in the Early Detection of Systemic Amyloidosis

The use of artificial intelligence (AI)-powered medical record data is effective in the early detection of systemic amyloidosis, according to a study being presented at the International Symposium on Amyloidosis 2024.

“Early detection of systemic amyloidosis is crucial for prognosis and quality of life. AI predictive models offer the potential for timely identification. Studies using medical records for diagnostic algorithms are lacking,” the researchers noted.

In this dynamic, retrospective, cohort study, researchers analyzed 961 patients (average age, 65 years), comprising 218 cases and 743 controls. Patients in the population of interest were all affiliated with the Medical Care Program at the Hospital Italiano de Buenos Aires. The researchers retrospectively collected and analyzed data for each case and their respective controls 6 months prior to diagnosis. The data included demographic, clinical, laboratory, and imaging information. Subsequently, the investigators noted that machine learning models, employing resampling techniques, were applied to training and testing cohorts with internal validation through cross-validation.

According to the results, the following predictive variables best identified amyloidosis: carpal tunnel syndrome history, arrhythmia, abnormalities in free light chains, and imaging. When assessing performance, the AI-based model demonstrated the best area under the curve (over 90%) consistently in both cohorts.

“Medical record features demonstrated strong performance in early amyloidosis identification using the developed AI-driven model, particularly with the random forest and decision tree algorithms,” the researchers concluded.