Overview of studies and ethical considerations related to AI applications in precision oncology
| Study | AI application | Key findings | Ethical considerations | Reference |
|---|---|---|---|---|
| Kelly et al. (2019) | Patient outcomes prediction | AI can predict patient outcomes such as hospital readmission and mortality | Need for transparency in AI models and consideration of potential biases or discrimination | [11] |
| Bi et al. (2019) | Patient monitoring and early detection of cancer recurrence | AI can improve the detection of cancer recurrence and enable early intervention | Need for regulation of AI in patient monitoring, protection of patient privacy and informed consent, and ethical considerations for patient autonomy and access to treatment | [12] |
| Dias and Torkamani (2019) | Genetic testing | AI can predict the risk of hereditary cancer based on genetic data | Need for transparency in AI models and protection of genetic data privacy | [13] |
| Mudgal and Das (2020) | Radiology imaging interpretation | AI outperformed radiologists in detecting cancer | Need for oversight and regulation of AI in radiology to ensure patient safety and protection from bias | [14] |
| Schwendicke et al. (2020) | Treatment planning and clinical decision-making | AI can improve treatment outcomes and reduce costs | Need for transparent and explainable AI models, protection of patient privacy, and consideration of ethical implications for patient autonomy | [15] |
| Reddy et al. (2020) | Clinical trials and drug development | AI can improve patient selection for clinical trials and accelerate drug development | Need for transparent and explainable AI models, protection of patient privacy and informed consent, and ethical considerations for equitable access to new treatments | [16] |
| Razzak et al. (2020) | Early cancer diagnosis | AI can detect cancer at an earlier stage than traditional methods | Need for data privacy and security to protect patient information and prevent misuse of data | [17] |
| Carter et al. (2020) | Risk prediction and screening | AI can improve the accuracy of breast cancer screening and risk prediction | Need for informed consent, privacy protection, and consideration of the potential harms of overdiagnosis | [18] |
| Huynh et al. (2020) | Tumor segmentation and radiotherapy planning | AI can improve the accuracy of tumor segmentation and radiotherapy planning | Need for clinical validation, transparency, and regulation to ensure patient safety | [19] |
| Hartl et al. (2021) | Development of precision medicine treatments | AI can identify novel drug targets and improve drug efficacy | Need for regulation of AI in drug development, protection of patient privacy and informed consent, and ethical considerations for drug pricing and access | [20] |
| Muller et al. (2021) | Personalized treatment recommendations | AI can identify effective treatment options based on genetic and clinical data | Need for informed consent and patient education to ensure understanding of AI-based recommendations | [21] |
| Delso et al. (2021) | Clinical trial design | AI can optimize clinical trial design and recruitment | Need for ethical considerations such as consent, privacy protection, and potential biases | [22] |
| Ahmad et al. (2021) | Pathology interpretation | AI can assist in pathology interpretation and reduce errors | Need for validation, transparency, and consideration of potential biases or errors | [23] |
| Alabi et al. (2021) | Prognosis prediction | AI can predict cancer prognosis and survival rates based on clinical and genomic data | Need for ethical guidelines and regulations for the use of AI in prognostic applications | [24] |
| Luk et al. (2022) | Predictive modeling for cancer diagnosis and risk stratification | AI can accurately predict cancer risk based on patient data | Need for transparent and explainable AI models, protection of patient privacy, and ethical considerations for informed consent and non-discrimination | [25] |
