Open Access

REVIEW

Machine Intelligence for Mental Health Diagnosis: A Systematic Review of Methods, Algorithms, and Key Challenges

Ravita Chahar, Ashutosh Kumar Dubey^*

Chitkara University School of Engineering and Technology, Chitkara University, Baddi, 174103, Himachal Pradesh, India

* Corresponding Author: Ashutosh Kumar Dubey. Email:

(This article belongs to the Special Issue: Advanced Medical Imaging Techniques Using Generative Artificial Intelligence)

Computers, Materials & Continua 2026, 86(1), 1-65. https://doi.org/10.32604/cmc.2025.066990

Received 23 April 2025; Accepted 17 September 2025; Issue published 10 November 2025

Abstract

Objective: The increasing global prevalence of mental health disorders highlights the urgent need for the development of innovative diagnostic methods. Conditions such as anxiety, depression, stress, bipolar disorder (BD), and autism spectrum disorder (ASD) frequently arise from the complex interplay of demographic, biological, and socioeconomic factors, resulting in aggravated symptoms. This review investigates machine intelligence approaches for the early detection and prediction of mental health conditions. Methods: The preferred reporting items for systematic reviews and meta-analyses (PRISMA) framework was employed to conduct a systematic review and analysis covering the period 2018 to 2025. The potential impact of machine intelligence methods was assessed by considering various strategies, hybridization of algorithms, tools, techniques, and datasets, and their applicability. Results: Through a systematic review of studies concentrating on the prediction and evaluation of mental disorders using machine intelligence algorithms, advancements, limitations, and gaps in current methodologies were highlighted. The datasets and tools utilized in these investigations were examined, offering a detailed overview of the status of computational models in understanding and diagnosing mental health disorders. Recent research indicated considerable improvements in diagnostic accuracy and treatment effectiveness, particularly for depression and anxiety, which have shown the greatest methodological diversity and notable advancements in machine intelligence. Conclusions: Despite these improvements, challenges persist, including the need for more diverse datasets, ethical issues surrounding data privacy and algorithmic bias, and obstacles to integrating these technologies into clinical settings. This synthesis emphasizes the transformative potential of machine intelligence in enhancing mental healthcare.

---

Keywords

---

Supplementary Material

Supplementary Material File

---