Masaru Nakamura

Hysteroscopic Photodynamic Diagnosis Using 5-Aminolevulinic Acid: A High-Sensitivity Diagnostic Method for Uterine Endometrial Malignant Diseases

To examine the diagnostic accuracy of hysteroscopic photodynamic diagnosis (PDD) using 5-aminolevulinic acid (5ALA) in patients with endometrial cancer and premalignant atypical endometrial hyperplasia. A single-center, open-label, exploratory intervention study. University Hospital in Japan. Thirty-four patients who underwent hysteroscopic resection in the Department of Obstetrics and Gynecology at Keio University Hospital. Patients were given 5ALA orally approximately 3 hours before surgery and underwent observation of the uterine cavity and endometrial biopsy using 5ALA-PDD during hysteroscopic resection. Specimens were diagnosed histopathologically and the diagnostic sensitivity and specificity of hysteroscopic 5ALA-PDD for malignancy in the uterine cavity was determined. Red (R), blue (B), and green (G) intensity values were determined from PDD images, and the relationships of histopathological diagnosis with these values were used to develop a model for objective diagnosis of uterine malignancy. Three patients were excluded from the study because of failure of the endoscope system. A total of 113 specimens were collected endoscopically. The sensitivity and specificity of 5ALA-PDD for diagnosis of malignancy in the uterine cavity were 93.8% and 51.9%, respectively. The R/B ratio in imaging analysis was highest in malignant lesions, followed by benign lesions and normal uterine tissue, with significant differences among these groups (p <.05). The R/B and G/B ratios were used in a formula for prediction of malignancy based on logistic regression and the area under the receiver operating characteristic curve for this formula was 0.838. At a formula cutoff value of 0.220, the sensitivity and specificity for diagnosis of malignant disease were 90.6% and 65.4%, respectively. To our knowledge, this is the first study of the diagnostic accuracy of 5ALA-PDD for malignancies in the uterine cavity. Hysteroscopic 5ALA-PDD had higher sensitivity and identifiability of lesions. These findings suggest that hysteroscopic 5ALA-PDD may be useful for diagnosis of minute lesions.

Development of a prognostic prediction support system for cervical intraepithelial neoplasia using artificial intelligence-based diagnosis

Human papillomavirus subtypes are predictive indicators of cervical intraepithelial neoplasia (CIN) progression. While colposcopy is also an essential part of cervical cancer prevention, its accuracy and reproducibility are limited because of subjective evaluation. This study aimed to develop an artificial intelligence (AI) algorithm that can accurately detect the optimal lesion associated with prognosis using colposcopic images of CIN2 patients by utilizing objective AI diagnosis. We identified colposcopic findings associated with the prognosis of patients with CIN2. We developed a convolutional neural network that can automatically detect the rate of high-grade lesions in the uterovaginal area in 12 segments. We finally evaluated the detection accuracy of our AI algorithm compared with the scores by multiple gynecologic oncologists. High-grade lesion occupancy in the uterovaginal area detected by senior colposcopists was significantly correlated with the prognosis of patients with CIN2. The detection rate for high-grade lesions in 12 segments of the uterovaginal area by the AI system was 62.1% for recall, and the overall correct response rate was 89.7%. Moreover, the percentage of high-grade lesions detected by the AI system was significantly correlated with the rate detected by multiple gynecologic senior oncologists (r=0.61). Our novel AI algorithm can accurately determine high-grade lesions associated with prognosis on colposcopic images, and these results provide an insight into the additional utility of colposcopy for the management of patients with CIN2.