Related Experiment Video
Updated: Jun 23, 2025
Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
Published on: October 27, 2023
Leveraging deep learning for detecting red blood cell morphological changes in blood films from children with severe
Ezer Moysis1, Biobele J Brown2,3,4, Wuraola Shokunbi3,5
1Department of Computer Science, Faculty of Engineering Sciences, University College London, London, UK.
Insights
Deep learning models can now identify severe malaria anaemia (SMA) by detecting altered red blood cells (RBCs) in peripheral blood films. This AI approach offers scalable and accurate diagnostics for this critical childhood illness.
Area of Science:
- Medical diagnostics
- Artificial intelligence in healthcare
- Parasitology
Background:
- Severe malaria anaemia (SMA) poses a significant threat to children under five in sub-Saharan Africa.
- The spleen's phagocytotic activity in SMA leads to distinct red blood cell (RBC) morphological changes.
- Manual assessment of peripheral blood films (PBFs) for SMA is not scalable and prone to variability.
Purpose of the Study:
- To develop and validate a deep learning model for the systematic and large-scale identification of SMA.
- To leverage morphological alterations in RBCs as indicators for SMA detection using artificial intelligence.
- To improve the diagnostic and prognostic evaluation of SMA.
Main Methods:
- A weakly supervised Multiple Instance Learning framework was employed to create the MILISMA model.
- MILISMA was trained to identify SMA by detecting morphologically altered RBCs in PBFs.
- Statistical analyses and visual examinations were used to validate the model's findings.
Main Results:
- The MILISMA model achieved an 83% classification accuracy (AUC 87%) for SMA detection.
- The model identified statistically significant morphological distinctions in RBCs associated with SMA (p < 0.01).
- Visual analyses confirmed unique morphological features of SMA-affected RBCs.
Conclusions:
- Deep learning, specifically MILISMA, offers a scalable solution for identifying SMA through RBC morphology.
- The model enhances understanding of SMA pathology by characterizing RBC alterations.
- This AI-driven approach has the potential to refine SMA diagnostics and prognostics.
Abstract:
In sub-Saharan Africa, acute-onset severe malaria anaemia (SMA) is a critical challenge, particularly affecting children under five. The acute drop in haematocrit in SMA is thought to be driven by an increased phagocytotic pathological process in the spleen, leading to the presence of distinct red blood cells (RBCs) with altered morphological characteristics. We hypothesized that these RBCs could be detected systematically and at scale in peripheral blood films (PBFs) by harnessing the capabilities of deep learning models. Assessment of PBFs by a microscopist does not scale for this task and is subject to variability. Here we introduce a deep learning model, leveraging a weakly supervised Multiple Instance Learning framework, to Identify SMA (MILISMA) through the presence of morphologically changed RBCs. MILISMA achieved a classification accuracy of 83% (receiver operating characteristic area under the curve [AUC] of 87%; precision-recall AUC of 76%). More importantly, MILISMA's capabilities extend to identifying statistically significant morphological distinctions (p < 0.01) in RBCs descriptors. Our findings are enriched by visual analyses, which underscore the unique morphological features of SMA-affected RBCs when compared to non-SMA cells. This model aided detection and characterization of RBC alterations could enhance the understanding of SMA's pathology and refine SMA diagnostic and prognostic evaluation processes at scale.
More Related Videos
12:06Analyzing Mitochondrial Morphology Through Simulation Supervised Learning
Published on: March 3, 2023
10:50Detection and Quantification of Plasmodium falciparum in Aqueous Red Blood Cells by Attenuated Total Reflection Infrared Spectroscopy and Multivariate Data Analysis
Published on: November 2, 2018