A novel approach based on computing with words for monitoring the heart failure patients

Developments in medical science have provided new ways in which care can be taken of people suffering from the risk of heart failure at reduced medical expenses, such as through wearable sensors. These are more efficient than traditional health monitoring methods such as in-person visits to medical practitioners, clinics, etc. Unfortunately, wearable sensors can measure quantitative parameters such as blood pressure and heart rate but not qualitative ones such as ease of respiration, pain, etc. The values of qualitative parameters are generally expressed by a sick person in the form of 'words'. In real life scenarios, medical experts suggest plausible medical tests/treatment to patients using their experience based on his/her feedback in terms of 'words'. In this paper, we propose a new approach, called heart monitoring through perceptual computing (HMT Per-C), that assesses the medical condition of a person (under the risk of heart failure) by processing user feedback in terms of 'words' and generates recommendations about the medical attention needed to be given to him/her. HMT Per-C is based on the technique of perceptual computing, which is a computing with words (CWW) technique that models 'words' using interval type-2 fuzzy sets. We have also compared the recommendations generated by perceptual computing with those generated by other CWW approaches viz., extension principle, symbolic method and 2-tuple. We have found that the extension principle, symbolic method and 2-tuple failed to give accurate results in 8%, 44% and 28% cases, respectively. Therefore, we believe that our proposed approach, HMT Per-C, is better, more user-friendly and close to real life scenarios. An outcome of the present work is the ready to use mobile app, “HMT Per-C”, that complements the data obtained from the devices like the oximeter but does not replace them. It can be downloaded freely from http://sau.ac.in/∼cilab/.