1 Department of Computer Science, Lamar University, Beaumont, Texas, USA. Email: mr.akash016@gmail.com | ORCID: https://orcid.org/0009-0005-4268-2833
2 Department of Information Technology, Washington University of Science and Technology, Virginia, USA. Email: mahfuzislamkhanjabed@gmail.com | ORCID: https://orcid.org/0009-0001-2141-2894
3 College of Business, Westcliff University, Irvine, California, USA. Email: sumisarker957@gmail.com | ORCID: https://orcid.org/0009-0005-7714-3012
4 Department of Software Engineering, University of Maryland, Adelphi, MD, USA. Email: ebicalhodocarmo@student.umgc.edu | ORCID: https://orcid.org/0009-0008-9219-0134
5 Master of Science in Information System Management, Stanton University, Los Angeles, California, USA. Email: islam23.sharmin@gmail.com
6* MBBS, China Three Gorges University, Yichang Central People's Hospital, Hubei, China (Corresponding Author). Email: drmomtazmitu@gmail.com | ORCID: https://orcid.org/0009-0004-9730-404X
7 School of Business, International American University, Los Angeles, California, USA. Email: mehedipritom7524@gmail.com | ORCID: https://orcid.org/0009-0004-5418-683X
8 Raj Soin College of Business, Wright State University, Dayton, Ohio, USA. Email: saimaaktershikha505@gmail.com
9 MSc in Applied Statistics and Data Science, The University of Texas at Arlington, Texas, USA. Email: mdahnaftajwar.kamal@mavs.uta.edu | ORCID: https://orcid.org/0009-0006-6444-207X
Background
Anthracyclines and trastuzumab cause cardiotoxicity as one of the key side effects of cancer treatment, which often causes an irreversible effect on the heart, resulting in decreased survival of cancer patients. The attraction of artificial intelligence (AI) and genomic data analytics provides an innovative solution for early-time detection and targeted prevention of cardiotoxicity in individuals. Nevertheless, no tested, multidimensional frameworks have been validated that integrate these technologies to be put into use in an anticipatory clinical manner. The objective of this study was to assess the statistical reliability, the validity, and predictive ability of AI-based prediction models of cardiotoxicity when applied by health care professionals and researchers.
Methods
A 30-item Likert-scale questionnaire containing 206 participants (oncologists, cardiologists, researchers, and data scientists) was chosen as a 30-item questionnaire with a quantitative and cross-sectional design. To evaluate normality, reliability, validity, and inferential relationships, the statistical analysis in SPSS was done. The various tests to be used were the Shapiro-Wilk/Kolmogorov-Smirnov normality test, Cronbach's Alpha reliability test, Kaiser Meyer Olkin (KMO) and Bartlett Test of Sphericity validity tests, t-test, ANOVA, Kruskal-Wallis, Chi-Square, Correlation, and Regression tests to evaluate the tests inferentially.
Results
The data showed normal distribution (p > 0.05), a high level of reliability (α = 0.90), and a satisfactory level of validity (KMO = 0.812; χ² = 1465.72, p < 0.001). Most of the inferential analyses indicated a significant group difference based on gender and profession (p < 0.05). The correlation indicated that there are positive and strong relationships between constructs (r = 0.680081), and the regression model revealed that without AI awareness (β = 0.465) and Genomic integration (β = 0.419), there is no relationship with clinical utility (R² = 0.764, p = 0.001).
Conclusion
The results support that AI and genomic integration offer a statistically significant and clinically meaningful model of predicting risk of cardiotoxicity in patients with cancer. The paper indicates that AI-based genomic models in precision cardio-oncology can be introduced to facilitate the early detection, specific prevention, and personalized treatment strategies.
Keywords: Cardiotoxicity, Artificial Intelligence (AI), genomics, Precision Cardio-Oncology, Predictive Modeling, Machine Learning, Clinical Utility, Reliability, Validity, Regression Analysis.
How to cite this article: Akash MMR, Jabed MIK, Sarker S, Carmo EB, Islam S, Mitu MAM, Pritom MH, Shikha SA, Kamal MAT. Cardiotoxicity Prediction Models in Cancer Patients Using Artificial Intelligence and Genomics. Int J Drug Deliv Technol. 2026;16(23s): 60-73. DOI: 10.25258/ijddt.16.23s.7
Source of support: Nil.
Conflict of interest: Nil.