Application of the sliding window method and Mask-RCNN method to nuclear recognition in oral cytology

Background

We aimed to develop an artificial intelligence (AI)-assisted oral cytology method, similar to cervical cytology. We focused on the detection of cell nuclei because the ratio of cell nuclei to cytoplasm increases with increasing cell malignancy. As an initial step in the development of AI-assisted cytology, we investigated two methods for the automatic detection of cell nuclei in blue-stained cells in cytopreparation images.

Methods

We evaluated the usefulness of the sliding window method (SWM) and mask region-based convolutional neural network (Mask-RCNN) in identifying the cell nuclei in oral cytopreparation images. Thirty cases of liquid-based oral cytology were analyzed. First, we performed the SWM by dividing each image into 96 × 96 pixels. Overall, 591 images with or without blue-stained cell nuclei were prepared as the training data and 197 as the test data (total: 1,576 images). Next, we performed the Mask-RCNN by preparing 130 images of Class II and III lesions and creating mask images showing cell regions based on these images.

Results

Using the SWM method, the highest detection rate for blue-stained cells in the evaluation group was 0.9314. For Mask-RCNN, 37 cell nuclei were identified, and 1 cell nucleus was identified as a non-nucleus after 40 epochs (error rate:0.027).

Conclusions

Mask-RCNN is more accurate than SWM in identifying the cell nuclei. If the blue-stained cell nuclei can be correctly identified automatically, the entire cell morphology can be grasped faster, and the diagnostic performance of cytology can be improved.

Clinicopathological evaluation of pre-operative chemoradiotherapy with S-1 as a treatment for locally advanced oral squamous cell carcinoma

The administration of pre-operative chemotherapy with S-1 and concurrent radiotherapy at a total dose of 30 Gy was clinicopathologically evaluated as a treatment for locally advanced oral squamous cell carcinoma (OSCC) in the present study. The participants comprised 81 patients with OSCC, consisting of 29 patients with stage II disease, 12 patients with stage III disease and 40 patients with stage IV disease. All patients received a total radiation dose of 30 Gy in daily fractions of 2 Gy, 5 times a week, for 3 weeks, and the patients were concurrently administered S-1 at a dose of 80-120 mg, twice daily, over 4 consecutive weeks. Radical surgery was performed in all cases at 2-6 weeks subsequent to the end of pre-operative chemoradiotherapy. The most common adverse event was oropharyngeal mucositis, but this was transient in all patients. No severe hematological or non-hematological toxicities were observed. The clinical and histopathological response rates were 70.4 and 75.3%, respectively. Post-operatively, local failure developed in 6 patients (7.4%) and neck failure developed in 2 patients (2.5%). Distant metastases were found in 7 patients (8.6%). The overall survival rate, disease-specific survival rate and locoregional control rate at 5 years were 87.7, 89.9 and 90.6%, respectively. Locoregional recurrence occurred more frequently in patients that demonstrated a poor histopathological response compared with patients that demonstrated a good response (P<0.01). These results indicate that pre-operative S-1 chemotherapy with radiotherapy at a total dose of 30 Gy is feasible and effective for patients with locally advanced OSCC, and that little or no histopathological response may be a risk factor for locoregional recurrence in this treatment.