(1) Background: Cardiac amyloidosis (CA) is a rare and complex condition with poor prognosis.\nWhile novel therapies improve outcomes, many affected individuals remain undiagnosed due to a\nlack of awareness among clinicians. This study was undertaken to develop an expert-independent\nmachine learning (ML) prediction model for CA relying on routinely determined laboratory parameters.\n(2) Methods: In a first step, we developed baseline linear models based on logistic regression.\nIn a second step, we used an ML algorithm based on gradient tree boosting to improve our linear\nprediction model, and to perform non-linear prediction. Then, we compared the performance of\nall diagnostic algorithms. All prediction models were developed on a training cohort, consisting\nof patients with proven CA (positive cases, n = 121) and amyloidosis-unrelated heart failure (HF)\npatients (negative cases, n = 415). Performances of all prediction models were evaluated on a separate\nprognostic validation cohort with 37 CA-positive and 124 CA-negative patients. (3) Results: Our best\nmodel, based on gradient-boosted ensembles of decision trees, achieved an area under the receiver\noperating characteristic curve (ROC AUC) score of 0.86, with sensitivity and specificity of 89.2%\nand 78.2%, respectively. The best linear model had an ROC AUC score of 0.75, with sensitivity and\nspecificity of 84.6 and 71.7, respectively. (4) Conclusions: Our work demonstrates that ML makes it\npossible to utilize basic laboratory parameters to generate a distinct CA-related HF profile compared\nwith CA-unrelated HF patients. This proof-of-concept study opens a potential new avenue in the\ndiagnostic workup of CA and may assist physicians in clinical reasoning.
Loading....