Machine learning and artificial intelligence in haematology

Shouval, Roni; Fein, Joshua; Savani, Bipin N.; Mohty, Mohamad; Nagler, Arnon

doi:10.1111/bjh.16915

Cited by 74 publications

(48 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…By combining functional data with genomics data, including mapping of acquired mutations within the drug target, and clinical features, modeling can be performed to develop machine learning algorithms that predict treatment outcome. 89,90 If successful, this would allow development of personalized treatment strategies for CLL, needed to overcome treatment resistance.…”

Section: Discussionmentioning

confidence: 99%

Overcoming resistance to targeted therapies in chronic lymphocytic leukemia

Skånland

Mato

2021

Blood Advances

View full text Add to dashboard Cite

Insight into the critical role of B-cell receptor signaling for the pathogenesis of chronic lymphocytic leukemia (CLL) led to the development of targeted therapies directed at key regulators of cell survival. Agents targeting B-cell lymphoma-2 protein, Bruton’s tyrosine kinase (BTK), and phosphatidylinositol 3-kinase are approved for treatment of CLL, and have significantly improved the disease management. Nevertheless, acquired resistance to the targeted therapies is a challenge still to be resolved. The mechanisms underlying resistance are becoming clearer, and include secondary mutations within the drug target and activation of bypass pathways. This knowledge has allowed development of strategies to prevent and overcome treatment resistance. Approaches to prevent resistance include targeting bypass mechanisms by combination therapies, temporally sequencing of therapies, improved clinical trial designs, and real-time monitoring of patient response. A rational design of drug sequencing may secure effective treatment options at the relapsed setting. Next-generation inhibitors and bispecific antibodies have the potential to overcome resistance to the BTK inhibitor ibrutinib. Immunotherapy, including chimeric antigen receptor-modified T-cell therapy, is explored for relapsed CLL. Here, recent advances that have contributed to the understanding of resistance to targeted therapies in CLL are discussed. Strategies for managing resistance are reviewed, including translational, real-world, and clinical perspectives.

show abstract

Section: Discussionmentioning

confidence: 99%

Overcoming resistance to targeted therapies in chronic lymphocytic leukemia

Skånland

Mato

2021

Blood Advances

View full text Add to dashboard Cite

show abstract

“…Machine learning tools for decision support have already been applied in the field of haematology [20,19,17]. However, they are essentially limited to predict a given disease onset from CBC data.…”

Section: Objectivementioning

confidence: 99%

Optimising Criteria for Manual Smear Review Following Automated Blood Count Analysis: A Machine Learning Approach

Avalos

Touchais

Henríquez-Henríquez

2021

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

The complete blood count (CBC) performed by automated haematology analysers is the most common clinical procedure in the world. Used for health checkup, diagnosis and patient follow-up, the CBC impacts the majority of medical decisions. If the analysis does not fit an expected setting, the laboratory staff manually reviews a blood smear, which is highly time-consuming. Criteria for reviewing CBCs are based on international consensus guidelines and locally adjusted to account for laboratory resources and populations characteristics. Our objective is to provide a clinical laboratory decision support tool to identify which CBC variables are linked to an increased risk of abnormal manual smear and at which threshold values. Thus, we treat criteria adjustment as a feature selection problem. We propose a cost-sensitive Lasso-penalised additive logistic regression combined with stability selection, adapted to the peculiarities of data and context: class-imbalance, categorisation of continuous predictors, required stability and enhanced interpretability. Using simulated and real CBC data, we show that our proposal is competitive in terms of predictive performance (compared to deep neural networks) and model selection performance (provided that there is sufficient data in the neighbourhood of the true thresholds). The R code is publicly available as an open source project.

show abstract

“…Furthermore, methods such as shrinkage techniques and machine learning algorithms capable of dealing with high-dimensional data (ie, the number of features is high relative to the number of patients) are needed. 9 Finally, with all humility, physicians must acknowledge that there is inherent uncertainty to prediction. It is unrealistic to expect that features at the beginning of a patient's journey or even at the time of transplantation will unambiguously determine his fate.…”

Section: Real-world Validation Prospective Analyses Of Implemented Decision-aide Dynamic Model Updating With Incorporation Of New Data Stmentioning

confidence: 99%