Scholarlink

Purpose <p>For patients with oral cavity cancer (OCC) and oropharyngeal cancer (OPC), the time between initial diagnosis and start of treatment can have a significant impact on treatment outcomes. This perspective piece looks at current and emerging artificial intelligence (AI) applications relevant to expediting the diagnosis of OCC/OPC, with particular attention given to how such technologies can actually affect treatment in a clinically relevant manner.</p> Methods <p>We conducted a focused narrative review of current and emerging AI applications relevant to OCC/OPC diagnosis. AI tools were mapped onto three key points in the diagnostic workflow: (1) initiation of contact between patient and provider, (2) clinical evaluation of malignant potential and decision on biopsy necessity, and (3) histopathologic assessment and determination of treatment need. For each stage, we evaluated the potential impact of AI on timeliness and consistency of care, along with practical limitations such as data requirements and integration with clinician judgment.</p> Results <p>AI applications have been developed across multiple stages of OCC/OPC diagnosis, including risk stratification based on clinical and demographic data, computer-assisted analysis of clinical photographs for lesion detection, and digital pathology tools for grading dysplasia and malignancy. Thoughtful deployment of these technologies may enable earlier identification of patients at risk, facilitate appropriate referral and biopsy, and help standardize histopathologic interpretation across providers and institutions. However, broad implementation is challenged by limited access to large, diverse, and well-annotated datasets, variability in imaging and biopsy techniques, uncertainty regarding generalizability of models across settings, and the need for transparent, interpretable outputs that clinicians can reliably incorporate into practice.</p> Conclusion <p>AI has substantial potential to enhance the timeliness, accuracy, and consistency of OCC/OPC diagnosis, but it should be viewed as an adjunct to, rather than a replacement for, clinician expertise. Realizing its benefit will require careful integration into existing workflows, ongoing validation in diverse clinical environments, and attention to data quality and equity to reduce, rather than exacerbate, disparities in oral cancer care.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Evolving Landscape for Artificial Intelligence in the Diagnosis and Histopathologic Assessment of Oral and Oropharyngeal Cancer

  • 期刊论文

  • 发表日期：

    2026年5月28日

查看全文
  • Betty Ben Dor,
  • Stephen Sonis

摘要

Purpose

For patients with oral cavity cancer (OCC) and oropharyngeal cancer (OPC), the time between initial diagnosis and start of treatment can have a significant impact on treatment outcomes. This perspective piece looks at current and emerging artificial intelligence (AI) applications relevant to expediting the diagnosis of OCC/OPC, with particular attention given to how such technologies can actually affect treatment in a clinically relevant manner.

Methods

We conducted a focused narrative review of current and emerging AI applications relevant to OCC/OPC diagnosis. AI tools were mapped onto three key points in the diagnostic workflow: (1) initiation of contact between patient and provider, (2) clinical evaluation of malignant potential and decision on biopsy necessity, and (3) histopathologic assessment and determination of treatment need. For each stage, we evaluated the potential impact of AI on timeliness and consistency of care, along with practical limitations such as data requirements and integration with clinician judgment.

Results

AI applications have been developed across multiple stages of OCC/OPC diagnosis, including risk stratification based on clinical and demographic data, computer-assisted analysis of clinical photographs for lesion detection, and digital pathology tools for grading dysplasia and malignancy. Thoughtful deployment of these technologies may enable earlier identification of patients at risk, facilitate appropriate referral and biopsy, and help standardize histopathologic interpretation across providers and institutions. However, broad implementation is challenged by limited access to large, diverse, and well-annotated datasets, variability in imaging and biopsy techniques, uncertainty regarding generalizability of models across settings, and the need for transparent, interpretable outputs that clinicians can reliably incorporate into practice.

Conclusion

AI has substantial potential to enhance the timeliness, accuracy, and consistency of OCC/OPC diagnosis, but it should be viewed as an adjunct to, rather than a replacement for, clinician expertise. Realizing its benefit will require careful integration into existing workflows, ongoing validation in diverse clinical environments, and attention to data quality and equity to reduce, rather than exacerbate, disparities in oral cancer care.