An eXplainable Artificial Intelligence analysis of Raman spectra for thyroid cancer diagnosis

Abstract

Raman spectroscopy shows great potential as a diagnostic tool for thyroid cancer due to its ability to detect biochemical changes during cancer development. This technique is particularly valuable because it is non-invasive and label/dye-free. Compared to molecular tests, Raman spectroscopy analyses can more effectively discriminate malignant features, thus reducing unnecessary surgeries. However, one major hurdle to using Raman spectroscopy as a diagnostic tool is the identification of significant patterns and peaks. In this study, we propose a Machine Learning procedure to discriminate healthy/benign versus malignant nodules that produces interpretable results. We collect Raman spectra obtained from histological samples, select a set of peaks with a data-driven and label independent approach and train the algorithms with the relative prominence of the peaks in the selected set. The performance of the considered models, quantified by area under the Receiver Operating Characteristic curve, exceeds 0.9. To enhance the interpretability of the results, we employ eXplainable Artificial Intelligence and compute the contribution of each feature to the prediction of each sample.

Unraveling the microbiome-metabolome nexus: a comprehensive study protocol for personalized management of Behçet’s disease using explainable artificial intelligence

“…The algorithm learns to make predictions or decisions based on input data by generalizing from the labeled examples it has seen during training. Moreover, the XAI algorithm “SHapley Additive exPlanations” (SHAP) will be used to detect for each patient, which features are more important for the ML algorithm in its classification ( Bellantuono et al, 2023 ; Novielli et al, 2023 ). SHAP is an algorithm used in machine learning to explain the predictions made by complex models, particularly for models like XGBoost, Random Forest, neural networks, and others.…”

“…Feature Reduction : Feature reduction is a crucial aspect of our approach. Techniques such as Principal Component Analysis (PCA) ( Song et al, 2010 ) and wrapper methods like Boruta ( Kursa et al, 2010 ; Bellantuono et al, 2023 ) will be employed. These methods effectively reduce the dimensionality of the feature space, allowing us to train models even with a limited number of instances.…”

Explainable artificial intelligence for microbiome data analysis in colorectal cancer biomarker identification

“…This framework intricately involves data preprocessing, feature extraction, feature selection, and model construction, employing an array of ML algorithms. To ensure transparency and interpretability in our study, we embrace the principles of Explainable Artificial Intelligence (XAI) (Lombardi et al, 2021a , b ; Bellantuono et al, 2023 ; Novielli et al, 2023 ). XAI not only enhances the trustworthiness of our models but also empowers clinicians to understand the rationale behind each prediction.…”

Identification of Myocardial Infarction (MI) Probability from Imbalanced Medical Survey Data: An Artificial Neural Network (ANN) with Explainable AI (XAI) Insights

“…Despite the erroneous complexity of the data, the model was still able to yield very reliable outcomes. Interpretability 41 and sensitivity 42 analysis approaches were employed to ensure that the model’s predictions were explainable and therefore most practically useful in medical practice 43 , thus showing the distinct contributions of each input domain to the predictive framework.…”