This study suggests a novel Acute Lymphoblastic Leukemia (ALL) diagnostic model, built solely on complete blood count (CBC) records. Using a dataset comprised of CBC records of 86 ALL and 86 control patients respectively, we identified the most ALL-specific parameters using a feature selection approach. Next, Grid Search-based hyperparameter tuning with a five-fold cross-validation scheme was adopted to build classifiers using Random Forest, XGBoost, and Decision Tree algorithms. A comparison between the performances of the three models demonstrates that Decision Tree classifier outperformed XGBoost and Random Forest algorithms in ALL detection using CBC-based records.

Download full-text PDF

Source
http://dx.doi.org/10.3233/SHTI230479DOI Listing

Publication Analysis

Top Keywords

acute lymphoblastic
8
lymphoblastic leukemia
8
cbc records
8
random forest
8
decision tree
8
artificial intelligence-based
4
intelligence-based diagnostic
4
diagnostic system
4
system acute
4
leukemia detection
4

Similar Publications

Adolescent and young adult (AYA) survivors of acute lymphoblastic or myeloid leukemia diagnosed between the ages of 15 and 39 years are at risk for adverse late health effects following cancer treatment and require ongoing survivorship care. This study aims to understand the landscape of transitioning AYAs with leukemia from active treatment to survivorship care. A cross-sectional, anonymous online survey was sent out via listserv/email.

View Article and Find Full Text PDF

Unlabelled: Biochemical analyses of cerebrospinal fluid (CSF) are routinely performed at diagnosis in many pediatric oncology and hematology centers when acute leukemia is diagnosed. However, the clinical relevance of these analyses remains unclear. We conducted a retrospective analysis of biochemical CSF data from children diagnosed with acute leukemia at two French hospitals between 2016 and 2023 assessing the results in relation to the presence or absence of leukemic neuromeningeal involvement and the correlation between cytological and biochemical analyses.

View Article and Find Full Text PDF

Antibody-drug conjugates (ADCs) combining monoclonal antibodies with cytotoxic payloads are a rapidly emerging class of immune-based therapeutics with the potential to improve the treatment of cancer, including children with relapse/refractory acute lymphoblastic leukemia (ALL). CD123, the α subunit of the interleukin-3 receptor, is overexpressed in ALL and is a potential therapeutic target. Here, we show that pivekimab sunirine (PVEK), a recently developed ADC comprising the CD123-targeting antibody, G4723A, and the cytotoxic payload, DGN549, was highly effective in vivo against a large panel of pediatric ALL patient-derived xenograft (PDX) models ( = 39).

View Article and Find Full Text PDF

Background: This study aimed to develop a prognostic model based on extracellular trap-related genes (NETRGs) for patients with cALL.

Methods: Data from the TARGET-ALL-P2 and TARGET-ALL-P3 cohorts in the Genomic Data Commons database, the transcriptome dataset GSE26713, the single-cell transcriptome dataset GSE130116 from the Gene Expression Omnibus database and 306 NETRGs identified were analysed. Differentially expressed genes (DEGs) were identified from GSE26713 and differentially expressed NETRGs (DE-NETRGs) were obtained by overlapping DEGs with NETRGs.

View Article and Find Full Text PDF

Background: Killer immunoglobulin-like receptors (KIRs) are key molecules used by natural killer (NK) cells to interact with target cells. These receptors exhibit extensive genotypic polymorphism which has been associated with varying outcomes in immune responses against diseases. This study aimed to investigate the relationships between genotypes and haplotypes with acute lymphoblastic leukemia (ALL) in Saudi patients.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!