Deep learning enhances acute lymphoblastic leukemia diagnosis and classification using bone marrow images.

Acute lymphoblastic leukemia (ALL) poses a significant health challenge, particularly in pediatric cases, requiring precise and rapid diagnostic approaches. This comprehensive review explores the transformative capacity of deep learning (DL) in enhancing ALL diagnosis and classification, focusing on bone marrow image analysis. Examining ten studies conducted between 2013 and 2023 across various countries, including India, China, KSA, and Mexico, the synthesis underscores the adaptability and proficiency of DL methodologies in detecting leukemia. Innovative DL models, notably Convolutional Neural Networks (CNNs) with Cat-Boosting, XG-Boosting, and Transfer Learning techniques, demonstrate notable approaches. Some models achieve outstanding accuracy, with one CNN reaching 100% in cancer cell classification. The incorporation of novel algorithms like Cat-Swarm Optimization and specialized CNN architectures contributes to superior classification accuracy. Performance metrics highlight these achievements, with models consistently outperforming traditional diagnostic methods. For instance, a CNN with Cat-Boosting attains 100% accuracy, while others hover around 99%, showcasing DL models' robustness in ALL diagnosis. Despite acknowledged challenges, such as the need for larger and more diverse datasets, these findings underscore DL's transformative potential in reshaping leukemia diagnostics. The high numerical accuracies accentuate a promising trajectory toward more efficient and accurate ALL diagnosis in clinical settings, prompting ongoing research to address challenges and refine DL models for optimal clinical integration.

Canadian real-world experience of asciminib treatment in heavily pre-treated chronic myeloid leukemia (CML) patients who failed multiple lines of tyrosine kinase inhibitor (TKI) therapy.

BACKGROUND: Asciminib is a novel drug specifically targeting ABL myristoyl pocket in the ABL1 protein. METHODS: Forty one patients with chronic myeloid leukemia treated with asciminib from 2018 to 2022 were reviewed and analyzed for the efficacy and tolerability of asciminib using real-world experience data. RESULTS: The median age was 60 years (range 17-90) with a past history of a cardiovascular event in 21 patients (51%). Patients were pretreated with a median of 3 previous tyrosine kinase inhibitors (range 1-5). After a median of 12 months of asciminib (range 3-41), major molecular response (MMR) rate was 39% (n = 11/28) and 42% (n = 5/12) at 6 and 12 months, respectively. Molecular response with 2 log reduction (MR2) was noted in 54% (n = 15/28) and 50% (n = 6/12) at 6 and 12 months. The cumulative incidence of MMR and MR2 was 46.3% and 66% at 12 months. Five patients discontinued asciminib due to treatment failure (n = 3) or thrombocytopenia (n = 2). There were no cardiovascular events. Out of 7 patients treated with high dose asciminib for T315I mutation, 5 patients achieved MMR or deeper response. The event-free survival was 63% at 12 months. CONCLUSION: This study confirmed clinical efficacy and tolerability of asciminib with real-world experience.

Integrating AI and ML in Myelodysplastic Syndrome Diagnosis: State-of-the-Art and Future Prospects.

Myelodysplastic syndrome (MDS) is composed of diverse hematological malignancies caused by dysfunctional stem cells, leading to abnormal hematopoiesis and cytopenia. Approximately 30% of MDS cases progress to acute myeloid leukemia (AML), a more aggressive disease. Early detection is crucial to intervene before MDS progresses to AML. The current diagnostic process for MDS involves analyzing peripheral blood smear (PBS), bone marrow sample (BMS), and flow cytometry (FC) data, along with clinical patient information, which is labor-intensive and time-consuming. Recent advancements in machine learning offer an opportunity for faster, automated, and accurate diagnosis of MDS. In this review, we aim to provide an overview of the current applications of AI in the diagnosis of MDS and highlight their advantages, disadvantages, and performance metrics.

Reducing use of coagulation tests in a family medicine practice setting: An implementation study.

Introduction: Clinicians often order the international normalized ratio (INR) and activated partial thromboplastin time (APTT) to evaluate for the possibility of inherited bleeding disorders despite sensitivities and specificities of 1%-2%. The most accurate tool to evaluate for bleeding disorders is a validated bleeding assessment tool (BAT). Our aim was to reduce coagulation testing by >50% in a large family practice in Ontario, Canada. Methods: We conducted an implementation study from May 2016 to February 2020. Iterative interventions included introduction of a validated BAT into the electronic medical record (EMR); removal of the APTT as a prepopulated selection from the laboratory requisition; and education targeting family medicine teams and laboratory personnel. The primary outcome was the rate of pre- and post-APTT testing. Creatinine testing was the control. Data were analyzed via an interrupted time series analysis using Stata 13. Results: Immediately following education of the laboratory personnel on coagulation testing, the APTT rate level dropped by 1.26 tests per 100 patient visits per month (p < 0.001) and was sustained until the end of the study. Meanwhile, the PT/INR and creatinine testing rate levels did not change (rate level = -0.02 per 100 visits per month, p = 0.79 and 0.49, p = 0.22 respectively). There was good uptake of the BAT following integration and 18/88 (20%) obtained a referral to hematology after BAT completion. Conclusions: Multidisciplinary, iterative interventions reduced APTT testing and enabled the use of BATs to guide hematology referrals in a large family practice.

Lymphoma-associated acquired von Willebrand syndrome responsive to splenectomy: A case report.

A previously healthy 33-year-old female presented with a large hematoma over her right knee after kneeling. She was found to have pancytopenia and massive splenomegaly. Von Willebrand Factor (VWF) antigen level was 0.38 units/ml, ristocetin cofactor activity 0.13 units/ml, and VWF multimeric distribution was normal. Bone marrow examination revealed an indolent B-cell lymphoma. Diagnosis was consistent with acquired von Willebrand syndrome as an autoimmune epiphenomenon of a lymphoma. Diagnostic and therapeutic splenectomy under hemostatic coverage was performed. VWF antigen levels and activities immediately normalized postoperatively and remained within the normal range several months later. Splenic pathology confirmed hairy cell leukemia with a BRAF mutation.

The effects of non-pharmaceutical interventions on SARS-CoV-2 transmission in different socioeconomic populations in Kuwait: a modeling study.

BACKGROUND: Aggressive non-pharmaceutical interventions (NPIs) may reduce transmission of SARS-CoV-2. The extent to which these interventions are successful in stopping the spread have not been characterized in countries with distinct socioeconomic groups. We compared the effects of a partial lockdown on disease transmission among Kuwaitis (P1) and non-Kuwaitis (P2) living in Kuwait. METHODS: We fit a modified metapopulation SEIR transmission model to reported cases stratified by two groups to estimate the impact of a partial lockdown on the effective reproduction number ([Formula: see text]). We estimated the basic reproduction number ([Formula: see text]) for the transmission in each group and simulated the potential trajectories of an outbreak from the first recorded case of community transmission until 12 days after the partial lockdown. We estimated [Formula: see text] values of both groups before and after the partial curfew, simulated the effect of these values on the epidemic curves and explored a range of cross-transmission scenarios. RESULTS: We estimate [Formula: see text] at 1·08 (95% CI: 1·00-1·26) for P1 and 2·36 (2·03-2·71) for P2. On March 22nd, [Formula: see text] for P1 and P2 are estimated at 1·19 (1·04-1·34) and 1·75 (1·26-2·11) respectively. After the partial curfew had taken effect, [Formula: see text] for P1 dropped modestly to 1·05 (0·82-1·26) but almost doubled for P2 to 2·89 (2·30-3·70). Our simulated epidemic trajectories show that the partial curfew measure greatly reduced and delayed the height of the peak in P1, yet significantly elevated and hastened the peak in P2. Modest cross-transmission between P1 and P2 greatly elevated the height of the peak in P1 and brought it forward in time closer to the peak of P2. CONCLUSION: Our results indicate and quantify how the same lockdown intervention can accentuate disease transmission in some subpopulations while potentially controlling it in others. Any such control may further become compromised in the presence of cross-transmission between subpopulations. Future interventions and policies need to be sensitive to socioeconomic and health disparities.

Improving quality of care for patients with iron deficiency anemia presenting to the emergency department.

BACKGROUND: Patients presenting to the emergency department (ED) with iron deficiency anemia (IDA) are underrecognized, undertreated with iron, and overtransfused. A 3-month audit of red blood cell (RBC) transfusions at the Sunnybrook Health Sciences Centre ED in 2013 showed that only 53% of transfusions for IDA were appropriate. The aim of this quality improvement project was to increase the rate of appropriate transfusion to greater than 80%. STUDY DESIGN AND METHODS: Since November 2013, several quality improvement interventions have been implemented, including educational presentations, development of an algorithm on IDA management in the ED, and development of an ED IDA toolkit. The primary outcome was appropriateness of RBC transfusions per month. The process measure was monthly intravenous (IV) iron use in IDA patients managed exclusively by ED staff. Balancing measures included IV iron use according to the algorithm and undertransfusion. RESULTS: Over a 24-month period beginning January 2014, assessment of 193 units transfused in the ED showed an improvement of RBC appropriateness to 91% (range 50%-100%). IV iron use increased from one dose in the 3-month audit to an average of 2.6 and 4.7 per month in 2014 and 2015, respectively. IV iron use did not follow the algorithm in 19% (18 of 93) of cases: 12 were given to patients with less severe iron deficiency or bleeding. CONCLUSION: Improved RBC transfusion appropriateness for IDA in the ED can be achieved and maintained with the implementation of simple educational and practical interventions.