Explainable AI-driven decision support system for personalizing rehabilitation routines in stroke recovery

 Artificial intelligence is completely useless until we get EXACT 100% RECOVERY PROTOCOLS CREATED! Are you that blitheringly stupid you can't see that? You're using the tyranny of low expectations to ignore what survivors want!

You'll want 100% recovery when you are the 1 in 4 per WHO that has a stroke!

Since the goal of all stroke survivors is 100% recovery this is putting the cart before the horse! What fucking stupidity!

Send me hate mail on this: oc1dean@gmail.com. I'll print your complete statement with your name and my response in my blog. Or are you afraid to engage with my stroke-addled mind? I would like to know why you aren't solving stroke to 100% recovery, and what is your definition of competence in stroke? Swearing at me is allowed, I'll return the favor.

Explainable AI-driven decision support system for personalizing rehabilitation routines in stroke recovery

• Sergio Martínez-Cid,
• David Vallejo,
• Vanesa Herrera,
• Santiago Schez-Sobrino,
• José J. Castro-Schez &
• Javier A. Albusac 

• 2 Accesses

• Explore all metrics

Abstract

The technological revolution driven by advancements in Artificial Intelligence (AI) is radically transforming various sectors, with healthcare among the most positively impacted. This article is situated within the context of such transformation, highlighting the contribution of AI in supporting professionals dedicated to the physical rehabilitation of stroke survivors. Our study focuses on the design of a Decision Support System (DSS) integrated within a comprehensive remote rehabilitation framework, consisting of two interconnected applications: one for the therapist, designed to define routines and monitor patients, and another for the patient, enabling autonomous rehabilitation exercises at home. This DSS employs fuzzy logic, significantly enhancing its scalability and interpretability. We propose a system capable of automatically suggesting personalized adjustments to a patient’s rehabilitation routine based on their performance. Our approach can offer physiotherapists considerable time savings by automating routine adjustments, thereby allowing them to allocate more attention to personalized patient care and complex case analysis. Furthermore, this system incorporates principles of Artificial Intelligence (XAI), providing justifications for its suggestions. This affords therapists a stronger basis for validating or rejecting the proposed modifications by the artificial system. The paper presents a case study where a stroke patient’s rehabilitation routine is automatically adjusted by the system, demonstrating the applicability and benefits of our approach. The routine generated by the artificial system is compared with the routine that a physiotherapist could potentially assign and modify manually when monitoring the progress of a stroke patient. Finally, the findings of a preliminary evaluation with patients and therapists in a hospital are also discussed.