Ryo Kurokawa

Differentiation between ovarian metastasis from colorectal carcinoma and primary ovarian carcinoma: Evaluation of tumour markers and “mille-feuille sign” on computed tomography/magnetic resonance imaging

The purpose of this retrospective study was to evaluate the usefulness of serum tumour markers and morphological characteristics in CT/MRI to differentiate between ovarian metastases from colorectal carcinomas (OMCRC) and primary ovarian carcinomas (POC). Preoperative radiological images of 41 OMCRCs from 27 patients (mean age ± SD: 52.2 ± 10.7 years) and 46 POCs from 36 patients (52.1 ± 12.7 years) were included. Three blinded gynecological radiologists classified tumour morphology into 'mille-feuille sign', 'solid and cystic', 'multicystic without nodules', and 'multicystic with nodules' groups and analysed using Fisher's exact test. Serum carcinoembryonic antigen (CEA), cancer antigen 125 (CA125), and carbohydrate antigen 19-9 levels were compared by Wilcoxon rank-sum test. 'Mille-feuille sign' indicated OMCRC (OMCRC: 8/41, POC: 1/46, specificity = 0.98, p = 0.011) and had excellent interobserver agreement (Fleiss's kappa value = 0.96). 'Solid and cystic' indicated POC (18/41 vs 41/45, p < 0.001) and 'multicystic without nodules' indicated OMCRC (8/41 vs 2/46, p = 0.041). There was no significant difference in 'multicystic with nodules'. CA125 levels were higher in POCs (292.5 U/mL vs. 41.0 U/mL, p = 0.003). CEA levels were higher in OMCRCs (24.5 ng/mL vs 2 ng/mL, p < 0.001). CEA (< 6.3 ng/mL) AND (CA125 (≥87.0 U/mL) OR 'solid and cystic') indicated POC with high accuracy (3/41 vs 44/46, accuracy = 0.94, p < 0.001). Our new method with morphological classification and tumour markers were useful for differentiating the two tumours. In particular, the 'mille-feuille sign' frequently indicated OMCRC with high specificity and excellent interobserver agreement.

The automatic diagnosis artificial intelligence system for preoperative magnetic resonance imaging of uterine sarcoma

Magnetic resonance imaging (MRI) is efficient for the diagnosis of preoperative uterine sarcoma; however, misdiagnoses may occur. In this study, we developed a new artificial intelligence (AI) system to overcome the limitations of requiring specialists to manually process datasets and a large amount of computer resources. The AI system comprises a tumor image filter, which extracts MRI slices containing tumors, and sarcoma evaluator, which diagnoses uterine sarcomas. We used 15 types of MRI patient sequences to train deep neural network (DNN) models used by tumor filter and sarcoma evaluator with 8 cross-validation sets. We implemented tumor filter and sarcoma evaluator using ensemble prediction technique with 9 DNN models. Ten tumor filters and sarcoma evaluator sets were developed to evaluate fluctuation accuracy. Finally, AutoDiag-AI was used to evaluate the new validation dataset, including 8 cases of sarcomas and 24 leiomyomas. Tumor image filter and sarcoma evaluator accuracies were 92.68% and 90.50%, respectively. AutoDiag-AI with the original dataset accuracy was 89.32%, with 90.47% sensitivity and 88.95% specificity, whereas AutoDiag-AI with the new validation dataset accuracy was 92.44%, with 92.25% sensitivity and 92.50% specificity. Our newly established AI system automatically extracts tumor sites from MRI images and diagnoses them as uterine sarcomas without human intervention. Its accuracy is comparable to that of a radiologist. With further validation, the system could be applied for diagnosis of other diseases. Further improvement of the system's accuracy may enable its clinical application in the future.