Deep learning detects acute myeloid leukemia and predicts NPM1 mutation status from bone marrow smears

Abstract

The evaluation of bone marrow morphology by experienced hematopathologists is essential in the diagnosis of acute myeloid leukemia (AML); however, it suffers from a lack of standardization and inter-observer variability. Deep learning (DL) can process medical image data and provides data-driven class predictions. Here, we apply a multi-step DL approach to automatically segment cells from bone marrow images, distinguish between AML samples and healthy controls with an area under the receiver operating characteristic (AUROC) of 0.9699, and predict the mutation status of Nucleophosmin 1 (NPM1)—one of the most common mutations in AML—with an AUROC of 0.92 using only image data from bone marrow smears. Utilizing occlusion sensitivity maps, we observed so far unreported morphologic cell features such as a pattern of condensed chromatin and perinuclear lightening zones in myeloblasts of NPM1-mutated AML and prominent nucleoli in wild-type NPM1 AML enabling the DL model to provide accurate class predictions.

Efficient and Highly Accurate Diagnosis of Malignant Hematological Diseases Based on Whole-Slide Images Using Deep Learning

“…We found that using only slide-level labeling without detailed pixellevel or cellular annotation enables interpretable, highperformance diagnostics, which overcomes the cost of labeling and closely resembles clinical applications. We collected a large collection of bone marrow smear WSIs and used them for training, which has more information than training a model using an expert-selected ROI (24,26,45) and allows for full automation. Simultaneously, we demonstrate that no adjustments are required and can be applied to the micrograph analysis of bone marrow or peripheral blood, which addresses the high imaging cost in certain resource-poor regions.…”

Shedding Light on the Black Box of a Neural Network Used to Detect Prostate Cancer in Whole Slide Images by Occlusion-Based Explainability

“…This should be considered when planning similar studies. The occlusion-based explanations pointing to critical structures represent a useful tool for fine-tuning and optimization of neural networks in histopathology, and potentially for identification of previously unrecognized morphological features related to histopathological diagnosis, prognosis, and prediction [23, 24]. Finally, unravelling the large quantity of features within the network and exposing the key elements will help to promote trust in these and similar AI-based methods in pathology, enhancing the opportunities for incorporation into clinical use.…”

Computational analysis of peripheral blood smears detects disease-associated cytomorphologies

“…Recent applications of computational cytomorphology on bone marrow smears have demonstrated its ability to automatically identify different leukocytes 30,62 and assist diagnostic predictions [27][28][29] in specialized haemato-oncology. By demonstrating that this can now be extended to blood smears/slides, our work reveals the potential for the large-scale incorporation of automated cytomorphology into routine diagnostic workflows.…”

“…Computational methods, which have shown promise in the characterization and prediction of MDS and AML using bone marrow slides [27][28][29] and identification of abnormal leukocytes 30 , can help address these problems.…”