Analysis of fuzzy based intelligent health care application system for the diagnosis of mental health in women with ovarian cancer using computational models
Affiliations: [a] Department of Computer Science and Automation, Indian Institute of Science, Bangalore, India | [b] Department of Artificial Intelligence and Data Science, Global Academy of Technology, Bangalore, India
Correspondence:
[*]
Corresponding author: Susheela Devi, Department of Computer Science and Automation, Indian Institute of Science, Bangalore, India. %****␣idt-17-idt228006_temp.tex␣Line␣50␣**** E-mail: kashwini@iisc.ac.in.
Abstract: Depending on the characteristics of the cancer and the specific treatment required, each type of cancer comes with a unique set of challenges in the psychological wellbeing of women. This research work mainly focuses on Ovarian cancer since the it is one of the 5th leading cancers among women. As per the statistics of 2021, by the American Cancer Society, 21,410 women would be diagnosed with ovarian cancer and 13,770 women might die from ovarian cancer. Both physically and psychologically, ovarian cancer presents several challenges. To control the growth of the tumour, multiple treatments are required. The psychological issues in women with ovarian cancer is mainly due to “loss of femininity” that affects them while they proceed through the phases of diagnosis, treatment and recurrence. Psychological factors associated with both, having ovarian cancer and being at risk are considered in this study. In the proposed work, PHQ 9 and GAD 7 are the tools used to measure depression and anxiety among women who are undergoing treatment for ovarian cancer. The data, collected with the help of these tools, is analysed using the popular Machine Learning algorithms such as k-Nearest Neighbour (kNN), Random Forest, Support Vector Machine (SVM), Artificial Neural Network etc. The results of Machine Learning algorithms are then compared with Mamdani and Sugeno fuzzy inference models. The Sugeno fuzzy inference system outperformed in comparison to all other models, with an accuracy of 96.2% for depression and 98.83% for anxiety, followed by Mamdani fuzzy inference system giving 94.3% accuracy for depression and 96.7% for anxiety. The performance is then compared with the linear SVM which obtained an accuracy of 91.52% for depression and 93.78% for anxiety. The analysed performance of the data using computational algorithms are compared with that of expert clinical psychologists. The severely affected women are advised to visit a psychiatrist.
