Single Shot Multibox Detector Automatic Polyp Detection Network Based on Gastrointestinal Endoscopic Images.

PURPOSE: In order to resolve the situation of high missed diagnosis rate and high misdiagnosis rate of the pathological analysis of the gastrointestinal endoscopic images by experts, we propose an automatic polyp detection algorithm based on Single Shot Multibox Detector (SSD). METHOD: In the paper, SSD is based on VGG-16, the fully connected layer is changed to a convolutional layer, and four convolutional layers with successively decreasing scales are added as a new network structure. In order to verify the practicability, it is not only compared with manual polyp detection but also with Mask R-CNN. RESULTS: Multiple experimental results show that the mean Average Precision (mAP) of the SSD network is 95.74%, which is 12.4% higher than the manual detection and 5.7% higher than the Mask R-CNN. When detecting a single frame of image, the detection speed of SSD is 8.41 times that of manual detection. CONCLUSION: Based on the traditional pattern recognition algorithm and the target detection algorithm using deep learning, we select a variety of algorithms to identify and classify polyps to achieve efficient detection results. Our research demonstrates that deep learning has a lot of room for development in the field of gastrointestinal image recognition.

Relations between brain network activation and analgesic effect induced by low vs. high frequency electrical acupoint stimulation in different subjects: a functional magnetic resonance imaging study.

Two- or 100-Hz electrical acupoint stimulation (EAS) can induce analgesia via distinct central mechanisms. It has long been known that the extent of EAS analgesia showed tremendous difference among subjects. Functional MRI (fMRI) studies were performed to allocate the possible mechanisms underlying the frequency specificity as well as individual variability of EAS analgesia. In either frequencies, the averaged fMRI activation levels of bilateral secondary somatosensory area and insula, contralateral anterior cingulate cortex and thalamus were positively correlated with the EAS-induced analgesic effect across the subjects. In 2-Hz EAS group, positive correlations were observed in contralateral primary motor area, supplementary motor area, and ipsilateral superior temporal gyrus, while negative correlations were found in bilateral hippocampus. In 100-Hz EAS group, positive correlations were observed in contralateral inferior parietal lobule, ipsilateral anterior cingulate cortex, nucleus accumbens, and pons, while negative correlation was detected in contralateral amygdala. These results suggest that functional activities of certain brain areas might be correlated with the effect of EAS-induced analgesia, in a frequency-dependent dynamic. EAS-induced analgesia with low and high frequencies seems to be mediated by different, though overlapped, brain networks. The differential activations/de-activations in brain networks across subjects may provide a neurobiological explanation for the mechanisms of the induction and the individual variability of analgesic effect induced by EAS, or that of manual acupuncture as well.