Study on artificial intelligence-based ultrasonic-assisted diagnosis for developmental dysplasia of the hip / 中华骨科杂志

Methods:Two thousand standard sections images werre collected from 2 000 clinical retrospective pediatric hip ultrasound videos from January 2019 to January 2021. All standard sections were annotated by the annotation team through the self-designed software based on Python 3.6 environment for image cross-media data annotation and manual review standardization process with unified standards. Among them, 1 753 were randomly selected for training the deep learning system, and the remaining 247 were used for testing the system. Further, 200 standard sections were randomly selected from the test set, and 8 clinicians independently completed the film reading annotation. The 8 independent results were then compared with the AI results.Results:The testing set consists of 247 patients. Compared with the clinician's measurements, the area under the receiver operating characteristic curve (AUC) of diagnosing hip joint maturity was 0.865, the sensitivity was 76.19%, and the specificity was 96.9%. The AUC of AI system interpretation under Graf detailed typing was 0.575, the sensitivity was 25.90%, the specificity was 89.10%. The 95% LoA of α-angle determined by Bland-Altman method, of -4.7051° to 6.5948° ( Bias -0.94, P<0.001), compared with clinicians' measurements. The 95% LoA of β-angle, of -7.7191 to 6.8777 ( Bias -0.42, P=0.077). Compared with those from 8 clinicians, the results of AI system interpretation were more stable, and the β-angle effect was more prominent. Conclusion:The AI system can quickly and accurately measure the Graf correlation index of standard DDH ultrasonic standard diagnosis plane.

Research on characteristics of resting-state functional connectivity strengths in patients with amnestic mild cognitive impairment / 中华放射学杂志

Objective To explore a new index for reflecting the topological information of brain functional networks in patients at high risk of Alzheimer disease using characteristics of resting-state functional connectivity strengths(FCS) in patients with amnestic mild cognitive impairment(aMCI). Methods Thirty-one aMCI patients and 42 age, gender and years of education matched normal controls were enrolled between September 2009 and April 2011 in this study. The resting-state functional MRI (rs-fMRI) data of all participants were acquired and preprocessed. Then the whole-brain functional connectivities were constructed for exploring the distribution characteristics of hub regions which had higher FCS values. Using two-sample t test to compare group differences in age, years of education and each neuropsychological assessment. In addition, using Chi-squared test to compare group differences in gender. Group differences in FCS values were analyzed by general linear model. Finally, correlation analyses were used to evaluate the relationships between the FCS values of the brain regions with group differences and behavioral scores in aMCI patients. Results The hub regions of the functional networks in the aMCI patients were mainly located in the association cortices such as the precuneuses, posterior cingulate cortices, medial prefrontal cortices, angular gyri, superior occipital gyri, fusiform gyri and lingual gyri. The distribution models in the aMCI patients were consistent with those in the normal controls. However, the FCS values of these brain regions were significantly lower in the aMCI patients than those in the normal controls. In comparison to the normal controls, the aMCI patients had significantly decreased FCS values in the bilateral fusiform gyri, lingual gyri, superior occipital gyri, left middle occipital gyrus and postcentral gyrus (the cluster was 389, 230, 187 and 107 voxels, respectively;P<0.05, respectively), and they had decreased trends of FCS values in the bilateral posterior cingulate cortices and right insulas. The correlation analysis with uncorrected conditions showed that the FCS values of the left postcentral gyri were correlatid with the clock drawing test (CDT) scores (r=0.436, P=0.026). Conclusions aMCI mainly attacks the hub regions of brain functional networks. The changes of functional connectivities in aMCI may reflect the early pathophysiologic alterations of AD.

Clinical application of image processing and neural network in cytopathological diagnosis of lung cancer / 中华胸心血管外科杂志

Objective To study a new technique (lung cancer diagnossing system, LCDS) based on the computer imaging and artificial neural network for early diagnosis of lung cancer, and evaluate it's value in early cytopathological diagnosis of lung cancer. Methods The cytological smears from the specimens obtained by Percutaneous Aspiration Lung Biopsy (PALB) in 512 cases were synthetically analyzed by LCDS. Among them, 362 cases received operations. The diagnoses by LCDS were compared with postoperative histopathological diagnosis. Results In cytopathological diagnoses for the 512 specimens, LCDS can judge between cancer cells and non-cancer cells from lung lesions with its image analysis and expert system. Moreover, it can distinguish squamous carcinoma, adenocarcinoma and small cell carcinoma in cytopathological diagnosis with built-in neural network. The total coincident rate of LCDS diagnosis was 91.80% compared with the pathological diagnosis. In the 362 cases, the sensitivity of LCDS diagnosis was 94.79% (291/307), the specificity was 90.91%(50/55), and the consistent rate was 94.20%(341/362). Conclusion The diagnostic pattern of LCDS was practical and effective. It has applicable value in cytopathological diagnosis of lung cancer and may be an efficient means for early diagnosis of lung cancer.

Prediction on lengths of stay in the postanesthesia care unit following general anesthesia: preliminary study of the neural network and logistic regression delling

The length of stay in the postanesthesia care unit (PACU) following general anesthesia in adults is an important issue. A model, which can predict the results of PACU stays, could improve the utilization of PACU and operating room resources through a more efficient arrangement. The purpose of study was to compare the performance of neural network to logistic regression analysis using clinical sets of data from adult patients undergoing general anesthesia. An artificial neural network was trained with 409 clinical sets using backward error propagation and validated through independent testing of 183 records. Twenty-two inputs were used to find determinants and to predict categorical values. Logistic regression analysis was performed to provide a comparison. The neural network correctly predicted in 81.4% of situations and identified discriminating variables (intubated state, sex, neuromuscular blocker and intraoperative use of opioid), whereas the figure was 65.0% in logistic regression analysis. We concluded that the neural network could provide a useful predictive model for the optimization of limited resources. The neural network is a new alternative classifying method for developing a predictive paradigm, and it has a higher classifying performance compared to the logistic regression model.