Research progress of artificial intelligence in the early screening, diagnosis, precise treatment and prognosis prediction of three central gynecological malignancies

Abstract

Early screening and diagnosis, precise treatment and prognosis prediction are pivotal for enhancing cancer patients' survival rates and for improving their life quality, it is the same case in gynecological malignancies. Among gynecological malignancies, endometrial cancer, cervical cancer and ovarian cancer are regarded as the three major types due to their high incidence rates, clinical severity and unique biological characteristics. Nowadays, the early screening and diagnosis of gynecological malignancies mainly depends on imaging examinations, pathological evaluations, and serum biomarkers assessments, while they possess inherent limitations. The treatment and prognosis prediction of gynecological malignant tumors show significant individualized differences. Although the treatment methods have been continuously improved, there are still shortcomings such as complex drug resistance mechanisms that limit the treatment effect, the impact of treatment toxicity on the quality of life of patients, and the impact of varying doctor experience levels across hospitals on disparities in diagnosis and treatment quality. With the rapid evolution of artificial intelligence (AI), particularly through the integration of deep learning (DL) and machine learning (ML) algorithms, AI technologies have shown their benefits in medicine. AI technologies can efficiently analyze medical images, genomic data, and clinical information, thereby enable more precise diagnoses, facilitate the design of personalized treatment strategies, predict treatment outcomes and recurrence risks. AI has also been utilized in gynecological malignancies and exhibited substantial potential. This review summarizes the latest advancements of AI applications in the early screening, diagnosis, treatment and prognosis prediction of three central gynecological malignancies and dialectically discussed current limitations. It provides valuable insights into the future translational potential of AI in gynecological oncology.

Ähnliche Arbeiten