ABSTRACT
OBJECTIVE: To explore the clinical application value of artificial intelligence image aided diagnosis platformbased on Faster R-CNN in identifying EMVI of rectal cancer.METHODS: In the multicenter retrospective study,500 patients with rectal cancer who underwent high-resolution MRI examination between July 2016 and February 2019 wereselected from seven hospitals in China. They were divided into 174 positive and 326 negative patients. Patients wererandomized to a training group(400 patients,including 133 positive and 267 negative) and a validation group(100 patients,including 41 positive and 59 negative) using a random number method. Using the Faster R-CNN to learn and train 20 430 high-resolution MRI images of thetraining group,an artificial intelligence image-aided diagnosis platform was established. The5107 high-resolution MRI images of thevalidation group were clinically validated.Receiver operating characteristic(ROC) curveand area under the curve(AUC) were used tocompare the diagnostic results of the artificialintelligence image-aided diagnosis platform andthe senior image expert.RESULTS: The accuracy,sensitivity,specificity,positive predictive valueand negative predictive value of EMVI forartificial intelligence image-aided diagnosis platform were 93.4%, 97.3%, 89.5%, 0.90 and 0.97,respectively. The area under the receiver operating characteristiccurve(AUC) was 0.98. The time required to automatically recognize a single image was 0.2 seconds,which had clearadvantages compared to radiologists(estimated to be about 10 seconds).CONCLUSION: The artificial intelligence image-assisted diagnosis platform based on Faster R-CNN has high efficiency and feasibility for identifying rectal cancerEMVI,and can assist imaging diagnosis.
ABSTRACT
BACKGROUND@#Artificial intelligence-assisted image recognition technology is currently able to detect the target area of an image and fetch information to make classifications according to target features. This study aimed to use deep neural networks for computed tomography (CT) diagnosis of perigastric metastatic lymph nodes (PGMLNs) to simulate the recognition of lymph nodes by radiologists, and to acquire more accurate identification results.@*METHODS@#A total of 1371 images of suspected lymph node metastasis from enhanced abdominal CT scans were identified and labeled by radiologists and were used with 18,780 original images for faster region-based convolutional neural networks (FR-CNN) deep learning. The identification results of 6000 random CT images from 100 gastric cancer patients by the FR-CNN were compared with results obtained from radiologists in terms of their identification accuracy. Similarly, 1004 CT images with metastatic lymph nodes that had been post-operatively confirmed by pathological examination and 11,340 original images were used in the identification and learning processes described above. The same 6000 gastric cancer CT images were used for the verification, according to which the diagnosis results were analyzed.@*RESULTS@#In the initial group, precision-recall curves were generated based on the precision rates, the recall rates of nodule classes of the training set and the validation set; the mean average precision (mAP) value was 0.5019. To verify the results of the initial learning group, the receiver operating characteristic curves was generated, and the corresponding area under the curve (AUC) value was calculated as 0.8995. After the second phase of precise learning, all the indicators were improved, and the mAP and AUC values were 0.7801 and 0.9541, respectively.@*CONCLUSION@#Through deep learning, FR-CNN achieved high judgment effectiveness and recognition accuracy for CT diagnosis of PGMLNs.@*TRIAL REGISTRATION@#Chinese Clinical Trial Registry, No. ChiCTR1800016787; http://www.chictr.org.cn/showproj.aspx?proj=28515.