A human-interpretable machine learning pipeline based on ultrasound to support leiomyosarcoma diagnosis.

The preoperative evaluation of myometrial tumors is essential to avoid delayed treatment and to establish the appropriate surgical approach. Specifically, the differential diagnosis of leiomyosarcoma (LMS) is particularly challenging due to the overlapping of clinical, laboratory and ultrasound features between fibroids and LMS. In this work, we present a human-interpretable machine learning (ML) pipeline to support the preoperative differential diagnosis of LMS from leiomyomas, based on both clinical data and gynecological ultrasound assessment of 68 patients (8 with LMS diagnosis). The pipeline provides the following novel contributions: (i) end-users have been involved both in the definition of the ML tasks and in the evaluation of the overall approach; (ii) clinical specialists get a full understanding of both the decision-making mechanisms of the ML algorithms and the impact of the features on each automatic decision. Moreover, the proposed pipeline addresses some of the problems concerning both the imbalance of the two classes by analyzing and selecting the best combination of the synthetic oversampling strategy of the minority class and the classification algorithm among different choices, and the explainability of the features at global and local levels. The results show very high performance of the best strategy (AUC = 0.99, F1 = 0.87) and the strong and stable impact of two ultrasound-based features (i.e., tumor borders and consistency of the lesions). Furthermore, the SHAP algorithm was exploited to quantify the impact of the features at the local level and a specific module was developed to provide a template-based natural language (NL) translation of the explanations for enhancing their interpretability and fostering the use of ML in the clinical setting.

Clinical decision support system for end-stage kidney disease risk estimation in IgA nephropathy patients.

BACKGROUND: The progression of IgA nephropathy (IgAN) to end-stage kidney disease (ESKD) depends on several factors that are not quite clear and tangle the risk assessment. We aimed at developing a clinical decision support system (CDSS) for a quantitative risk assessment of ESKD and its timing using available clinical data at the time of renal biopsy. METHODS: We included a total of 1040 biopsy-proven IgAN patients with long-term follow-up from Italy (N = 546), Norway (N = 441) and Japan (N = 53). Of these, 241 patients reached ESKD: 104 Italian [median time to ESKD = 5 (3-9) years], 134 Norwegian [median time to ESKD = 6 (2-11) years] and 3 Japanese [median time to ESKD = 3 (2-12) years]. We independently trained and validated two cooperating artificial neural networks (ANNs) for predicting first the ESKD status and then the time to ESKD (defined as three categories: ≤ 3 years, between > 3 and 8 years and over 8 years). As inputs we used gender, age, histological grading, serum creatinine, 24-h proteinuria and hypertension at the time of renal biopsy. RESULTS: The ANNs demonstrated high performance for both the prediction of ESKD (with an AUC of 89.9, 93.3 and 100% in the Italian, Norwegian and Japanese IgAN population, respectively) and its timing (f-measure of 90.7% in the cohort from Italy and 70.8% in the one from Norway). We embedded the two ANNs in a CDSS available online (www.igan.net). Entering the clinical parameters at the time of renal biopsy, the CDSS returns as output the estimated risk and timing of ESKD for the patient. CONCLUSIONS: This CDSS provides useful additional information for identifying 'high-risk' IgAN patients and may help stratify them in the context of a personalized medicine approach.