The Applications of Machine Learning in the Management of Patients Undergoing Stem Cell Transplantation: Are We Ready?

Cancers, Journal Year: 2025, Volume and Issue: 17(3), P. 395 - 395

Published: Jan. 25, 2025

Hematopoietic stem cell transplantation (HSCT) is a life-saving therapy for hematologic malignancies, such as leukemia and lymphoma other severe conditions but associated with significant risks, including graft versus host disease (GVHD), relapse, treatment-related mortality. The increasing complexity of clinical, genomic, biomarker data has spurred interest in machine learning (ML), which emerged transformative tool to enhance decision-making optimize outcomes HSCT. This review examines the applications ML HSCT, focusing on donor selection, conditioning regimen, prediction post-transplant outcomes. Machine approaches, decision trees, random forests, neural networks, have demonstrated potential improving compatibility algorithms, mortality relapse prediction, GVHD risk stratification. Integrating “omics” models enabled identification novel biomarkers development highly accurate predictive tools, supporting personalized treatment strategies. Despite promising advancements, challenges persist, standardization, algorithm interpretability, ethical considerations regarding patient privacy. While holds promise revolutionizing HSCT management, addressing these barriers through multicenter collaborations regulatory frameworks remains essential broader clinical adoption. In addition, can cope some harmonization, patients’ protection, availability adequate infrastructure. Future research should prioritize larger datasets, multimodal integration, robust validation methods fully realize ML’s

Language: Английский

---

Application of artificial intelligence in chronic myeloid leukemia (CML) disease prediction and management: a scoping review

BMC Cancer, Journal Year: 2024, Volume and Issue: 24(1)

Published: Aug. 20, 2024

Navigating the complexity of chronic myeloid leukemia (CML) diagnosis and management poses significant challenges, including need for accurate prediction disease progression response to treatment. Artificial intelligence (AI) presents a transformative approach that enables development sophisticated predictive models personalized treatment strategies enhance early detection improve therapeutic interventions better patient outcomes.

Language: Английский

---

Applications of Artificial Intelligence in Acute Promyelocytic Leukemia: An Avenue of Opportunities? A Systematic Review

Journal of Clinical Medicine, Journal Year: 2025, Volume and Issue: 14(5), P. 1670 - 1670

Published: March 1, 2025

Background. Acute promyelocytic leukemia (APL) is a subtype of acute myeloid defined by the presence genetic abnormality, namely PML::RARA gene fusion, as result reciprocal balanced translocation between chromosome 17 and 15. APL veritable emergency in hematology due to risk early death coagulopathy if left untreated; thus, rapid diagnosis needed this hematological malignancy. Needless say, cytogenetic molecular biology techniques, i.e., fluorescent situ hybridization (FISH) polymerase chain reaction (PCR), are essential management patients diagnosed with APL. In recent years, use artificial intelligence (AI) its brances, machine learning (ML), deep (DL) field medicine, including hematology, has brought light new avenues for research fields blood cancers. However, our knowledge, there no comprehensive evaluation potential applications AI, ML, DL Thus, aim current publication was evaluate prospective uses these novel technologies Methods. We conducted literature search PubMed/MEDLINE, SCOPUS, Web Science identified 20 manuscripts eligible qualitative analysis. Results. The included publications highlight DL, other AI branches diagnosis, evaluation, examined models were based on routine biological parameters, cytomorphology, flow-cytometry and/or OMICS, demonstrated excellent performance metrics: sensitivity, specificity, accuracy, AUROC, others. Conclusions. can emerge relevant tool cases potentially contribute more screening identification emergency.

Language: Английский

---

[18F]FDG PET-Based Radiomics and Machine Learning for the Assessment of Gliomas and Glioblastomas: A Systematic Review

Information, Journal Year: 2025, Volume and Issue: 16(1), P. 58 - 58

Published: Jan. 16, 2025

Background: Some evidence of the value 18F-fluorodesoxyglucose ([18F]FDG) positron emission tomography (PET) imaging for assessment gliomas and glioblastomas (GBMs) is emerging. The aim this systematic review was to assess role [18F]FDG PET-based radiomics machine learning (ML) in evaluation these neoplasms. Methods: A wide literature search PubMed/MEDLINE, Scopus, Cochrane Library databases made find relevant published articles on ML GBMs. Results: Eight studies were included review. Signatures, including analysis ML, generally demonstrated a possible diagnostic different characteristics GBMs, such as methylation status O6-methylguanine-DNA methyltransferase (MGMT) promoter, isocitrate dehydrogenase (IDH) genotype, alpha thalassemia/mental retardation X-linked (ATRX) mutation status, proliferative activity, differential diagnosis with solitary brain metastases or primary central nervous system lymphoma, prognosis patients. Conclusion: Despite some intrinsic limitations affecting review, initial insights promising technologies GBMs are Validation preliminary findings multicentric needed translate approaches clinical setting.

Language: Английский

---

Beyond TKIs: Advancing Therapeutic Frontiers with Immunotherapy, Targeted Agents, and Combination Strategies in Resistant Chronic Myeloid Leukemia

Hemato, Journal Year: 2025, Volume and Issue: 6(1), P. 6 - 6

Published: March 11, 2025

Background: Chronic myeloid leukemia (CML) relates to the abnormal presence of Philadelphia chromosome, which originates production BCR-ABL1 fusion protein and therefore leads neoplastic transformation unregulated cell growth. The advent tyrosine kinase inhibitors (TKIs) has resulted in tremendous improvements CML scenarios; however, there are practical difficulties, especially considering late stages disease. This review examines recently developed strategies that intended increase efficiency treatment by overcoming TKI resistance. Methods: We performed a literature such databases as PubMed, Scopus, Web Science, Embase for last ten years. following keywords were used studies: ‘CML’, ‘TKI resistance’, ‘novel therapies’, ‘immunotherapy’, ‘targeted agents’, ‘combination therapies’. Only those studies included clinical trials preclinical across-the-board developmental programs attempt target tumor at multiple levels not just focus on basic first-line TKIs. Results: In patients who do respond TKIs, novel therapeutics encompass ponatinib, asciminib, CAR-T immunotherapy, BCL-2 mTOR inhibition conjunction with therapy. addresses both BCR-ABL1-dependent independent resistance mechanisms, increasing chance achieving deeper molecular response reduced toxicity. Nonetheless, they exhibit diverse characteristics regarding efficacy, safety, cost, quality life effects. Discussion: numerous challenges remain understanding mechanisms resistance, long-term efficacy medicines, ideal combinations attain optimal outcomes. Areas future research include search other patterns tailoring specific treatments patients, incorporating AI improve diagnosis monitoring. Conclusion: introduction therapeutic techniques into practice needs collaborative approach persistent dynamism new findings from research. Our analysis indicates posed resistant disease complex require further protocol development.

Language: Английский

---

Utilization of Machine Learning in the Prediction, Diagnosis, Prognosis, and Management of Chronic Myeloid Leukemia

International Journal of Molecular Sciences, Journal Year: 2025, Volume and Issue: 26(6), P. 2535 - 2535

Published: March 12, 2025

Chronic myeloid leukemia is a clonal hematologic disease characterized by the presence of Philadelphia chromosome and BCR::ABL1 fusion protein. Integrating different molecular, genetic, clinical, laboratory data would improve diagnostic, prognostic, predictive sensitivity chronic leukemia. However, without artificial intelligence support, managing such vast volume be impossible. Considering advancements growth in machine learning throughout years, several models algorithms have been proposed for management Here, we provide an overview recent research that used specific on patients with leukemia, highlighting potential benefits adopting therapeutic contexts as well its drawbacks. Our analysis demonstrated great advancing precision treatment CML through combination clinical genetic data, testing, learning. We can use these powerful instruments to unravel molecular spatial puzzles overcoming current obstacles. A new age patient-centered hematology care will ushered this, opening door improved diagnosis accuracy, sophisticated risk assessment, customized plans.

Language: Английский

---

Study of Expression of MST3 in Myeloid Leukaemia

Medical Sciences, Journal Year: 2025, Volume and Issue: 13(2), P. 33 - 33

Published: April 1, 2025

Myeloid leukaemia (ML) is a cancer that occurs by the accumulation of abnormally multiplied myeloid cells in bone marrow, peripheral blood, and other related tissue. MST3 gene GCK family has role apoptosis, along with cellular functions like differentiation, cell cycle, metabolism, others. Objectives: The objectives this study were to count RBCs WBCs, expression ML control samples, perform an silico correlation on KRAS NRAS genes. Methods: counting WBCs was carried out using hemacytometer, studied RT-PCR, GEPIA. Results: RBC WBC levels differed from levels, found be upregulated comparison controls, 2.90–8.65-fold change, significant p-value > 0.05. A positive also between genes, r value correlation. Conclusions: From study, it could deduced might have pathogenesis, but further research needed its progression disease.

Language: Английский

---

The biology of chronic myeloid leukemia: an overview of the new insights and biomarkers

Frontiers in Oncology, Journal Year: 2025, Volume and Issue: 15

Published: May 8, 2025

Chronic myeloid leukemia is one of the onco-hematologic diseases in which identification disease markers and therapeutic advances have been particularly impactful. Despite this, significant gaps remain our understanding pathogenesis, progression, mechanisms immune escape, resistance to standard therapies. Recently, technology biological knowledge drawn attention several promising areas research. Among these, leukemic stem cells, miRNAs, extracellular vesicles, additional BCR::ABL1 mutations, with particular reference ASXL1 gene, most extensively investigated. In this review we summarized critically commented main findings on these key topics over past 5 years, evaluating their potential impact patient management role development new strategies.

Language: Английский

---

The impact of next-generation sequencing for diagnosis and disease understanding of myeloid malignancies

Expert Review of Molecular Diagnostics, Journal Year: 2024, Volume and Issue: 24(7), P. 591 - 600

Published: July 2, 2024

Defining the chromosomal and molecular changes associated with myeloid neoplasms (MNs) optimizes clinical care through improved diagnosis, prognosis, treatment planning, patient monitoring. This review will concisely describe techniques used to profile MNs clinically today, descriptions of challenges emerging approaches that may soon become standard-of-care.

Language: Английский

---