Towards Trustworthy Predictions of Conversion from Mild Cognitive Impairment to Dementia: A Conformal Prediction Approach Telma Pereira 1( ✉ ) , Sandra Cardoso 2 , Dina Silva 2 , Alexandre de Mendonça 2 , Manuela Guerreiro 2 , and Sara C. Madeira 3 1 INESC-ID and Instituto Superior Técnico (IST), University of Lisbon, Lisbon, Portugal telma.pereira@ist.utl.pt 2 Laboratory of Neurosciences, Faculty of Medicine, Institute of Molecular Medicine, University of Lisbon, Lisbon, Portugal 3 LASIGE and Faculty of Sciences, University of Lisbon, Lisbon, Portugal Abstract. Predicting progression from a stage of Mild Cognitive Impairment to Alzheimer’s disease is a major pursuit in current dementia research. As a result, many prognostic models have emerged with the goal of supporting clinical deci‐ sions. Despite the efforts, the lack of a reliable assessment of the uncertainty of each prediction has hampered its application in practise. It is paramount for clini‐ cians to know how much they can rely upon the prediction made for a given patient, in order to adjust treatments to the patient based on that information. In this exploratory study, we evaluated the Conformal Prediction approach on the task of making predictions with precise levels of confidence. Conformal predic‐ tion showed promising results. Using high confidence levels have the drawback of leaving a large number of MCI patients without prognostic (the classifier is not confident enough to give a single class). When using forced predictions, conformal predictors achieved classification performances as good as standard classifiers, with the advantage of complementing each prediction with a confi‐ dence value. Keywords: Conformal predictors · Confidence estimation · Mild cognitive impairment · Alzheimer’s disease · Prognostic prediction 1 Introduction Alzheimer’s disease (AD) is a neurodegenerative disease, causing cognitive impairment, with devastating effect on patients and their families, and a huge socio-economic impact in modern societies. Nowadays, more than 30 million people suffer from AD worldwide and its prevalence is expected to triple by 2050 [1]. Mild Cognitive Impairment (MCI) is considered as a transitive stage between healthy aging and dementia [1], suggesting these patients as a group of singular interest to follow-up studies and interventions. In this context, studying the predictive value of MCI for the progression to dementia is a major challenge in current dementia research [2, 3]. © Springer International Publishing AG 2017 F. Fdez-Riverola et al. (eds.), 11th International Conference on Practical Applications of Computational Biology & Bioinformatics, Advances in Intelligent Systems and Computing 616, DOI 10.1007/978-3-319-60816-7_19