Amyloidosis refers to a range of medical conditions in which misshapen proteins accumulate in various organs and tissues, forming insoluble fibrils. Cardiac amyloidosis is frequently linked to the buildup of misfolded transthyretin (TTR) or immunoglobulin light chains (AL). Delayed diagnosis, due to lack of disease awareness, results in a poor prognosis, especially in patients with AL amyloidosis. Early identification is therefore a key factor to improve patient outcomes. This study investigates the use of supervised machine-learning algorithms to support clinicians in classifying amyloidosis and control subjects. The aim of this work is to foster model interpretability reporting the most important risk factors in predicting the presence of cardiac amyloidosis. We analyzed electronic health records (EHRs) of 418 participants acquired in a time window of 12 years as part of a case-control study conducted in Fondazione Toscana Gabriele Monasterio (Italy) clinical practice. This work paves the way for the creation of digital health solutions that can aid in amyloidosis screening. The effective handling, analysis, and interpretation of these solutions can have a transformative effect on modern healthcare, offering new opportunities for improved patient care.