Unsupervised corneal contour extraction algorithm with shared model for dynamic deformation videos: improving accuracy and noise resistance.

BACKGROUND: In this study, an automatic corneal contour extraction algorithm with a shared model is developed to extract contours from dynamic corneal videos containing noise, which improves the accuracy of corneal biomechanical evaluation and clinical diagnoses. The algorithm does not require manual labeling and completes the unsupervised semantic segmentation of each frame in corneal dynamic deformation videos based on a fully convolutional deep-learning network using corneal geometry and texture information. RESULTS: We included 1027 corneal videos at Tianjin Eye Hospital (Nankai University Affiliated Eye Hospital) from May 2020 to November 2021. The videos were obtained by the ultra-high-speed Scheimpflug camera, and then we used the shared model mechanism to accelerate the segmentation of corneal regions in videos, effectively resist noise, determine corneal regions based on shape factors, and finally achieve automatic and accurate extraction of corneal region contours. The Intersection over Union (IoU) of the extracted and real corneal contours using this algorithm reached 95%, and the average overlap error was 0.05, implying that the extracted corneal contour overlapped almost completely with the real contour. CONCLUSIONS: Compared to other algorithms, the method introduced in this study does not require manual annotation of corneal contour data in advance and can still extract accurate corneal contours from noisy corneal videos with good repeatability.

Localized Corneal Biomechanical Alteration Detected In Early Keratoconus Based on Corneal Deformation Using Artificial Intelligence.

PURPOSE: This study aimed to develop a novel method to diagnose early keratoconus by detecting localized corneal biomechanical changes based on dynamic deformation videos using machine learning. DESIGN: Diagnostic research study. METHODS: We included 917 corneal videos from the Tianjin Eye Hospital (Tianjin, China) and Shanxi Eye Hospital (Xi'an, China) from February 6, 2015, to August 25, 2022. Scheimpflug technology was used to obtain dynamic deformation videos under forced puffs of air. Fourteen new pixel-level biomechanical parameters were calculated based on a spline curve equation fitting by 115,200-pixel points from the corneal contour extracted from videos to characterize localized biomechanics. An ensemble learning model was developed, external validation was performed, and the diagnostic performance was compared with that of existing clinical diagnostic indices. The performance of the developed machine learning model was evaluated using precision, recall, F1 score, and the area under the receiver operating characteristic curve. RESULTS: The ensemble learning model successfully diagnosed early keratoconus (area under the curve = 0.9997) with 95.73% precision, 95.61% recall, and 95.50% F1 score in the sample set (n=802). External validation on an independent dataset (n=115) achieved 91.38% precision, 92.11% recall, and 91.18% F1 score. Diagnostic accuracy was significantly better than that of existing clinical diagnostic indices (from 86.28% to 93.36%, all P <0.01). CONCLUSIONS: Localized corneal biomechanical changes detected using dynamic deformation videos combined with machine learning algorithms were useful for diagnosing early keratoconus. Focusing on localized biomechanical changes may guide ophthalmologists, aiding the timely diagnosis of early keratoconus and benefiting the patient's vision.

Artificial Intelligence-Based Diagnostic Model for Detecting Keratoconus Using Videos of Corneal Force Deformation.

Purpose: To develop a novel method based on biomechanical parameters calculated from raw corneal dynamic deformation videos to quickly and accurately diagnose keratoconus using machine learning. Methods: The keratoconus group was included according to Rabinowitz's criteria, and the normal group included corneal refractive surgery candidates. Independent biomechanical parameters were calculated from dynamic corneal deformation videos. A novel neural network model was trained to diagnose keratoconus. Tenfold cross-validation was performed, and the sample set was divided into a training set for training, a validation set for parameter validation, and a testing set for performance evaluation. External validation was performed to evaluate the model's generalizability. Results: A novel intelligent diagnostic model for keratoconus based on a five-layer feedforward network was constructed by calculating four biomechanical characteristics, including time of the first applanation, deformation amplitude at the highest concavity, central corneal thickness, and radius at the highest concavity. The model was able to diagnose keratoconus with 99.6% accuracy, 99.3% sensitivity, 100% specificity, and 100% precision in the sample set (n = 276), and it achieved an accuracy of 98.7%, sensitivity of 97.4%, specificity of 100%, and precision of 100% in the external validation set (n = 78). Conclusions: In the absence of corneal topographic examination, rapid and accurate diagnosis of keratoconus is possible with the aid of machine learning. Our study provides a new potential approach and sheds light on the diagnosis of keratoconus from a purely corneal biomechanical perspective. Translational Relevance: Our findings could help improve the diagnosis of keratoconus based on corneal biomechanical properties.

Comparative study on clinical efficacy of different methods for the treatment of intramural aortic hematoma.

To explore the difference of curative effect between different treatment modalities, in order to provide reference for the treatment of aortic intramural hematoma (IMH). 168 patients with aortic intramural hematoma diagnosed and treated from January 2010 to July 2020 were selected in the Second Affiliated Hospital of Nanchang University. Among them, 48 patients were diagnosed with Stanford A aortic intramural hematoma and 120 were diagnosed with Stanford B aortic intramural hematoma. According to the therapeutic methods, patients were divided into conservative treatment group and endovascular treatment group (TEVAR). For endovascular treatment group, according to the different timing of surgery, can be divided into acute phase group (onset within 72 h) and non-acute phase group (time of onset > 72 h).The clinical data and follow-up data were collected and analyzed by variance analysis and χ2 test. There were 168 patients diagnosed with aortic intramural hematoma 39 of them were (81.25%) Stanford A aortic intramural hematoma patients with pleural or pericardial effusion. For patient with Stanford A aortic intramural hematoma, endovascular treatment was performed in 15 patients (31.2%), and 33 cases (68.8%) for conservative treatment. The average follow-up (24.9 ± 13.9) was months. There were 120 patients with Stanford type B aortic intramural hematoma (71.4%), 60 patients received endovascular treatment (50%), and 60 patients (50%) received conservative treatment, with an average follow-up of (27.8 ± 14.6) months. For Stanford A type aortic intramural hematoma patients when the maximum aortic diameter ≥ 50 mm or hematoma thickness ≥ 11 mm, with high morbidity and mortality, positive endovascular treatment can reduce complications and death. For patients with Stanford type B aortic intramural hematoma, when the maximum aortic diameter ≥ 40 mm or hematoma thickness ≥ 10 mm, with high morbidity and mortality, positive endovascular treatment can reduce complications and death. Both Stanford type A and B aortic intramural hematoma patients could benefit from the endovascular treatment when the initial maximum aortic diameter is ≥ 50 mm or the hematoma thickness is ≥ 11 mm.

Therapeutic Effect of Epidermal Growth Factor Combined With Nano Silver Dressing on Diabetic Foot Patients.

Objective: To investigate the clinical efficacy of epidermal growth factor combined with nano silver dressing in the treatment of diabetic foot wounds. Methods: A total of 160 patients with diabetic foot ulcers admitted to the Second Affiliated Hospital of Nanchang University from 2015-06 to 2018-06 were selected to participate in the experiment. A randomized table method was used to randomly divide 160 patients into 4 groups: 40 in the epidermal growth factor group, 40 in the nano-silver dressing group, 40 in the combined group, and 40 in the saline control group (normal saline). The healing stage of the wound surface and the growth degree of granulation tissue were graded. Each group was given a dressing change every other day, and the time required for wound repairing to each healing stage was observed. After 2 and 4 weeks of treatment, the wound exudate was collected for bacterial culture. Results: There was no significant difference in the time between the four groups of patients reaching the effective phase of treatment (level 1). Compared with the control group, the epidermal growth factor group and the combined group achieved a shorter time for wound repairing to healing stages 2 and 3, and the difference was significant (p < 0.05). The combined group had a shorter wound repairing time than the epidermal growth factor group (p < 0.05). Compared with the control group, the positive rate of bacteria in the combined group and the silver nanoparticles group was significantly lower after 2 and 4 weeks of treatment. Conclusion: There is no significant difference in wound healing between the four groups during the clinically effective period. After this period, the combined use of recombinant epidermis Growth factors and nano-silver dressings have a significant effect on promoting wound healing and can effectively prevent infection.

Autophagic regulation of cell growth by altered expression of Beclin 1 in triple-negative breast cancer.

Beclin 1 is a promoter gene for autophagy as well as a key factor for regulating tumor cell growth and death. Allelic deletion of Beclin 1 has been observed in certain triple-negative breat cancer (TNBC) cells, and it might be associated with increased proliferation and invasion in TNBC cells. In this study we investigated the relationship between Beclin 1 expression and prognosis for TNBC patients, as well as the influence on cell growth by Beclin 1 overexpression in different cultural conditions. Beclin 1 expression in TNBC tissues was measured by immunohistochemical staining and correlated with clinicopathologic parameters for TNBC patients. The plasmid of pDS-RED-C1-Beclin 1 was transfected to BT-549 and MDA-MB-231 cells and autophagy, proliferation, apoptosis, cell cycle and Epithelial-mesenchymal transition (EMT) process were measured. Results indicated that high level of Beclin 1 expression was correlated with more lymph nodes and distant metastasis but unrelated to survival rates in 5 years for TNBC patients. In vitro, overexpression of Beclin 1 improved cellular autophagy in both BT-549 and MDA-MB-231 cells, inhibited cell proliferation at normal cultural condition and increased cell survival in starvation, hypoxia or with doxorubicin stimulation. Besides, Beclin 1 overexpression decreased cell apoptosis, induced cells to be in G0/G1 phase and promoted EMT process through Wnt/ß-catenin pathway in starvation. Thus, Beclin 1 overexpression plays a double role in BT-549 and MDA-MB-231 cell growth by elevating the capability of autophagy. These findings might be useful for searching a proper method for clinical therapy of TNBC from the aspect of autophagy in future.

[Simultaneous analysis of theaflavins and catechins by capillary electrophoresis].

In order to monitor the changes of theaflavins and catechins during the process of tea polyphenol oxidation, a rapid, accurate, sensitive and convenient analytical method was established. The optimum analytical conditions for simultaneous analysis of four theaflavins (TFs) and six catechins by capillary electrophoresis were investigated. The electrolyte solution consisted of 200 mmol/L boric acid (pH 7.7), 10 mmol/L potassium dihydrogenphosphate, 9 mmol/L beta-cyclodextrin and 27.5% acetonitrile. The experimental conditions of the instrument were selected as follows: voltage 25 kV, column temperature 30 degrees C, and detection wavelength 200 nm. Ten constituents could be separated from each other completely within 8 min and each one had a good linear relationship between its peak area and corresponding concentration (nu = 0.990 7 - 0.999 8). Furthermore, their detection limits were from 0.39 to 0.88 microg/mL, the mean recoveries varied from 91.5% to 113.5%. The relative standard deviations were less than 5%. All the above results showed that the developed method is especially suitable for the analysis of theaflavins and catechins.