Table 4 SWOT analysis of the utilization of AI technology in the cancer screening program in the ASEAN region

INTERNAL FACTORS

STRENGTHS +

WEAKNESSES –

○ AI improves patient management by reducing unnecessary interventions and providing accurate diagnostic information [87].

○ AI paired with skilled specialists outperforms individual alone with recommended guidelines [88].

○ AI-assisted chest radiography enhances outreach to low-risk populations who are not screened and raises public screening uptake. [82].

○ AI integrated systems serve as a "second radiologist," assisting in interpreting X-ray images of breast cancer for prompt diagnosis and advancing research on decision-support systems for mammography [83].

○ In breast clinics, over-the-counter screening models for breast cancer enhance patient flow and reduce waiting times for appointments and investigations [86].

○ AI algorithms in smartphone applications identify and classify precancerous lesions of cervical cancer in real-time, eliminating time-consuming lab analysis [92, 93].

○ AI-based classification models help general practitioners in remote areas who use telemedicine for oral cancer screening [95].

○ Patients' and clinicians' inconsistent and mixed views regarding the adoption of AI can make integration difficult and necessitate stakeholder education [82].

○ Variability in the reproducibility of AI-assisted screening due to variations in the image acquisition protocols used by various devices [82].

○ Reliance on retrospective analysis; prospective studies are required to validate AI's efficacy and cost-effectiveness [84].

○ Algorithmic bias is a concern when relying on commercial AI software that was trained on specific regional datasets [87].

○ Use of AI in resource-limited settings with non-expert operators may increase false positives and unnecessary investigations [94].

○ Limited data for developing AI models [95].

EXTERNAL FACTORS

OPPORTUNITIES +

THREATS –

○ The use of AI in cancer screening makes early detection and intervention possible for underprivileged and remote populations [92, 93].

○ AI-integrated smartphone applications can reach remote areas and underserved populations, enhancing early detection and intervention [92, 93]

○ Limited research on AI's real effect in clinical practice, including recall for screening and individualized cancer screening needs to be explored [85].

○ Diverse guidance throughout the ASEAN region regarding AI governance in healthcare and ethical issues like accountability, transparency, privacy, and diagnostic bias [100].