Acute lymphoblastic leukemia cancer diagnosis in children and adults using transforming blood fluorescence microscopy imaging
Leukemia is a highly aggressive kind of cancer that may impact the bone marrow. The most fatal type acute lymphoblastic leukemia (ALL), is characterized by the excessive growth of immature white blood cells in the bone marrow. For diagnostic purposes, hematologists and experts use a state-of-the-art...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PeerJ Inc.
2025-08-01
|
| Series: | PeerJ Computer Science |
| Subjects: | |
| Online Access: | https://peerj.com/articles/cs-2997.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Leukemia is a highly aggressive kind of cancer that may impact the bone marrow. The most fatal type acute lymphoblastic leukemia (ALL), is characterized by the excessive growth of immature white blood cells in the bone marrow. For diagnostic purposes, hematologists and experts use a state-of-the-art microscope fitted with a high-powered magnifying lens to analyze blood and bone marrow samples. Experts attribute the rapid progress to the presence of adolescent white blood cells, not fully developed ones. A good treatment for ALL, no matter where it comes from, includes chemotherapy, medication given through a transplant. Experts have difficulties in accurately evaluating explosive cell features due to the onerous and time-consuming nature of manual diagnosis for this disease. A total of 89 individuals suspected of having ALL underwent sample collection, resulting in the acquisition of 3,256 images. The dataset is classified into four different types of cancer: early-stage, benign cells, pre-cancerous cells, and pro-cancer cells. The proposed approach employs several preprocessing and augmentation techniques to improve the results. The studies demonstrate that the technique achieved a recall rate of 100% for the pro-cell cancer subtype, an overall accuracy of 98.67% using enhanced data, and an overall accuracy of 97.87% using the original data. The experiments have shown that the proposed equipment is superior in reliability and accuracy compared to existing approaches, and it facilitates early detection in medical imaging. |
|---|---|
| ISSN: | 2376-5992 |