Related Experiment Video
Updated: Jan 9, 2026
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
7.3K
Comprehensive review on learning models of leukemia detection based on morphological information
Umarani Ponnusamy1, Viswanathan Perumal2
1School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, India.
Leukemia & Lymphoma
|December 4, 2025
Summary
Artificial intelligence (AI) aids in leukemia diagnosis by analyzing medical images. This systematic review explores AI methods for early and accurate detection, improving patient outcomes in blood cancer care.
Area of Science:
- Hematology
- Medical Imaging
- Artificial Intelligence
Background:
- Leukemia, the 11th most prevalent cancer globally, necessitates early and accurate diagnosis for effective treatment and improved patient survival.
- Conventional diagnostic methods like microscopy and blood counts are often time-consuming, unreliable, and require specialized expertise.
- Delayed or inaccurate diagnosis significantly increases disease progression, infection risk, and mortality rates.
Purpose of the Study:
- To systematically review artificial intelligence (AI)-based approaches for leukemia diagnosis.
- To analyze various image acquisition and preprocessing techniques used in AI-driven leukemia detection.
- To examine machine learning (ML) and deep learning (DL) models for leukemia classification, identifying challenges and future research directions.
Main Methods:
- Systematic literature review of AI applications in leukemia diagnosis.
- Analysis of image acquisition methods: peripheral blood smear microscopy, flow cytometry, bone marrow biopsy, and advanced imaging.
- Evaluation of preprocessing techniques: noise/artifact removal and image enhancement.
- Overview of segmentation algorithms (clustering to deep learning) and ML/DL classification models.
Main Results:
- AI approaches demonstrate satisfactory results in leukemia detection, offering improvements over conventional methods.
- The review covers diverse image analysis techniques, from basic preprocessing to advanced deep learning segmentation and classification.
- Various ML and DL models are discussed, with their respective advantages and limitations highlighted.
Conclusions:
- AI-powered tools show significant potential for enhancing the accuracy and timeliness of leukemia diagnosis.
- Further research is needed to address existing challenges and optimize AI models for clinical application in hematologic malignancies.
- AI integration can lead to better treatment decisions and improved survival rates for leukemia patients.
