# Accu-Help: A Machine Learning based Smart Healthcare Framework for Accurate Detection of Obsessive Compulsive Disorder

**Authors:** Kabita Patel, Ajaya Kumar Tripathy, Laxmi Narayan Padhy, Sujita Kumar Kar, Susanta Kumar Padhy, Saraju Prasad Mohanty  
**arXiv:** [2212.02346](https://arxiv.org/abs/2212.02346)  
**Date:** 2022-12-05  
**Summary page:** https://www.summarizepaper.com/en/paper/2212.02346v1/

## Key points

- The article discusses the development of a fully automated OCD detection model called Accu-Help.
- The proposed framework aims to detect OCD from oxidative stress biomarkers (OSBs) using machine learning.
- OSBs play a significant role in OCD individuals and can be measured through blood samples.
- The proposed approach involves segregating OSBs into three classes: Healthy Individual (HI), Genetically Affected Individual (GAI), and OCD Affected Individual (OAI).
- The OSBs considered for the study include superoxide dismutase (SD), Glutathione Peroxidase (GP), Catalase (CAT), Malondialdehyde (MAL), and serum cortisol (SC).
- To achieve high detection accuracy, the proposed framework involves distributed data collection, data integration, coordination between hospitals and biochemical laboratories, dynamic machine learning OCD prediction mode design using suitable algorithms such as neural networks-based approaches.
- The Accu-Help model has an OCD prediction accuracy of 86 percent.
- This approach offers an alternative mechanism for detecting OCD without requiring high-end machines such as EEG or MRI while achieving high detection accuracy.

## Summary

This article focuses on the development of a fully automated, smart, and accurate OCD detection conceptual model called Accu-Help. The proposed framework aims to help biochemical laboratories efficiently detect OCD from oxidative stress biomarkers (OSBs) using machine learning. The study suggests that oxidative stress biomarkers play a significant role in OCD individuals and can be measured through blood samples. The proposed approach involves segregating OSBs into three classes: Healthy Individual (HI), Genetically Affected Individual (GAI), and OCD Affected Individual (OAI). The OSBs considered for the study include superoxide dismutase (SD), Glutathione Peroxidase (GP), Catalase (CAT), Malondialdehyde (MAL), and serum cortisol (SC). To achieve high detection accuracy, the proposed framework involves distributed data collection, data integration, coordination between hospitals and biochemical laboratories, dynamic machine learning OCD prediction mode design using suitable algorithms such as neural networks-based approaches. The Accu-Help model has an OCD prediction accuracy of 86 percent. In summary, this article presents a novel contribution to the field by proposing a fully automated framework for efficient detection of OCD from OSBs using machine learning algorithms. This approach offers an alternative mechanism for detecting OCD without requiring high-end machines such as EEG or MRI while achieving high detection accuracy.