1B.tech, undergraduate student, SRM Institute of Technology, Ramapuram Campus, Chennai, India. Email: arvindhsb2005@gmail.com
2Professor, Department of Anaesthesiology, Mahatma Gandhi Medical College and Research Institute, Sri Balaji Vidyapeeth (Deemed to be University), Pondicherry, India - 607402. Email: painfreepartha@gmail.com; ORCID: 0000-0002-3808-6722
*Corresponding Author: S. Parthasarathy, Professor, Department of Anaesthesiology, Mahatma Gandhi Medical College and Research Institute, Sri Balaji Vidyapeeth (Deemed to be University), Pondicherry, India - 607402. Email: painfreepartha@gmail.com
Received: 19th Oct, 2025; Revised: 15th Dec, 2025; Accepted: 16th Jan, 2026; Available Online: 15th Feb, 2026
Anemia is one of the most significant global health concerns, with one-third of the world's population, including children and pregnant women, affected by it." The traditional method of diagnosing anemia involves blood tests, which can be expensive, time-consuming, and challenging, especially for developing countries. Recent developments in artificial intelligence, machine learning, and deep learning have enabled the creation of non-invasive diagnostic tools for detecting anemia, which utilize computer vision for the detection of biomarkers such as the color of tissues such as the conjunctiva, fingernails, palmar tissue, and gingiva, which indicate the presence of pallor, which is associated with hemoglobin deficiency. This scoping review aims to provide an overview of the literature on artificial intelligence-based computer vision for the detection of anemia, which has been conducted between 2017 and 2025. The literature indicates that there is high diagnostic accuracy, which is above 90%, for detecting anemia using convolutional neural networks, support vector machines, and transformer-based models, while image preprocessing plays a significant role in improving the performance of the model. The findings highlight the growing role of computer-vision and machine learning systems in enabling non-invasive screening of anemia through medical image analysis. Computer-based diagnostic platforms that incorporate smartphone imaging technologies may also assist in the development of rapid screening technologies in the community and primary care domains. However, some of the limitations of the current AI-based non-invasive screening technologies include the small datasets, differences in image acquisition conditions, and the lack of external validations in different populations. Despite the limitations, the AI-based non-invasive screening technologies hold significant promise to enhance the detection of anemia in resource-constrained environments.
Keywords: anemia, computer vision, medical images, artificial intelligence
How to cite this article: Arvindh SB, Parthasarathy S.., Computer-Vision–Based Non-Invasive Detection of Anemia from Medical Images: A Scoping Review of Artificial Intelligence with Applications in Drug Delivery and Therapeutic Monitoring..Int J Drug Deliv Technol. 2026; 16(2): 186-191; DOI: 10.25258/ijddt.16.2.23
Source of support: Nil.
Conflict of interest: None