29 November 2023

---

Artificial intelligence and oral cancer diagnosis

Lara Figini

---

Oral cancer (OC) is one of the most common forms of head and neck cancer and continues to have the lowest survival rates worldwide, despite advances in research and therapy. The prognosis of OC has not improved significantly in recent years, creating a persistent challenge in the biomedical field. According to the World Health Organization, it is estimated that there were 377,713 new cases and 177,757 deaths from lip and oral cavity cancer worldwide in 2020. It is estimated that over 90% of all oral cancers are oral squamous cell carcinomas (OSCC), highly aggressive and have a strong propensity to spread both locally and to other parts of the body. Early diagnosis of OSCC is vital for successful therapy, to increase the chances of survival, and to achieve lower rates of mortality and morbidity. Microscopy-based histopathological analysis of tissue samples is considered the gold standard for the diagnosis and classification of oral cancer. However, this approach can be slow and error-prone, limiting its clinical utility. In recent times, significant efforts have been invested in studying the potential of artificial intelligence (AI) to improve medical diagnosis. Machine learning (ML) techniques identify distinguishable patterns from existing data, and rely on human knowledge and efforts to distinguish their characteristics. In the field of oncology, artificial intelligence (AI) has seen rapid development, with notable successes reported in recent times.

Materials and methods

In a recent review, published in Biomedicines, the authors critically evaluated the studies available in the literature regarding the use of artificial intelligence in the diagnosis, classification and prediction of oral cancer (OC) using histopathological images. An electronic search of several databases, including PubMed, Scopus, Embase, Cochrane Library, Web of Science, Google Scholar and Saudi Digital Library was conducted for articles published on the topic between January 2000 and January 2023. Nineteen studies, which satisfied the inclusion criteria were subjected to critical analysis using QUADAS-2, and the certainty of the evidence was assessed using the GRADE approach.

Results

AI models have been widely applied in oral cancer diagnosis, differentiation between benign and malignant tumors, prediction of survival of OC patients, and OC grade. The AI models used in these studies showed an accuracy between 89.47% and 100%, a sensitivity between 97.76% and 99.26%, and a specificity between 92% and 99.42%. AI's ability to diagnose, classify and predict the occurrence of OC has been found to surpass existing clinical approaches.

Conclusions

From the data of this review, which must be confirmed in other similar studies and reviews, it can be concluded that artificial intelligence has excellent potential to offer superior results in terms of precision and accuracy, helping oral pathologists to significantly improve their diagnostic results and reduce the probability of error.

Info: Application and Performance of Artificial Intelligence (AI) in Oral Cancer Diagnosis and Prediction Using Histopathological Images: A Systematic Review. Sanjeev B. Khanagar, Lubna Alkadi, Maryam A. Alghilan, Sara Kalagi, Mohammed Awawdeh, Lalitytha Kumar Bijai, Satish Vishwanathaiah, Ali Aldhebaib and Oinam Gokulchandra Singh. Biomedicines. 2023 June 1;11(6):1612. doi: 10.3390/biomedicines11061612.