Objective To evaluate the performance of the artificial intelligence (AI)-enabled snail identification system for recognition of Oncomelania hupensis robertsoni and Tricula in schistosomiasis-endemic areas of Yunnan Province. Methods Fifty O. hupensis robertsoni and 50 Tricula samples were collected from Yongbei Township, Yongsheng County, Lijiang City, a schistosomiasis-endemic area in Yunnan Province in May 2024. A total of 100 snail sample images were captured with smartphones, including front-view images of 25 O. hupensis robertsoni and 25 Tricula samples (upward shell opening) and back-view images of 25 O. hupensis robertsoni and 25 Tricula samples (downward shell opening). Snail samples were identified as O. hupensis robertsoni or Tricula by schistosomiasis control experts with a deputy senior professional title and above according to image quality and morphological characteristics. A standard dataset for snail image classification was created, and served as a gold standard for recognition of snail samples. A total of 100 snail sample images were recognized with the AI-enabled intelligent snail identification system based on a WeChat mini program in smartphones. Schistosomiasis control professionals were randomly sampled from stations of schistosomisis prevention and control and centers for disease control and prevention in 18 schistosomiasis-endemic counties (districts, cities) of Yunnan Province, for artificial identification of 100 snail sample images. All professionals are assigned to two groups according the median years of snail survey experiences, and the effect of years of snail survey experiences on O. hupensis robertsoni sample image recognition was evaluated. A receiver operating characteristic (ROC) curve was plotted, and the sensitivity, specificity, accuracy, Youden’s index and the area under the curve (AUC) of the AI-enabled intelligent snail identification system and artificial identification were calculated for recognition of snail sample images. The snail sample image recognition results of AI-enabled intelligent snail identification system and artificial identification were compared with the gold standard, and the internal consistency of artificial identification results was evaluated with the Cronbach’s coefficient alpha. Results A total of 54 schistosomiasis control professionals were sampled for artificial identification of snail sample image recognition, with a response rate of 100% (54/54), and the accuracy, sensitivity, specificity, Youden’s index, and AUC of artificial identification were 90%, 86%, 94%, 0.80 and 0.90 for recognition of snail sample images, respectively. The overall Cronbach’s coefficient alpha of artificial identification was 0.768 for recognition of snail sample images, and the Cronbach’s coefficient alpha was 0.916 for recognition of O. hupensis robertsoni snail sample images and 0.925 for recognition of Tricula snail sample images. The overall accuracy of artificial identification was 90% for recognition of snail sample images, and there was no significant difference in the accuracy of artificial identification for recognition of O. hupensis robertsoni (86%) and Tricula snail sample images (94%) (χ² = 1.778, P > 0.05). There was no significant difference in the accuracy of artificial identification for recognition of snail sample images with upward (88%) and downward shell openings (92%) (χ² = 0.444, P > 0.05), and there was a significant difference in the accuracy of artificial identification for recognition of snail sample images between schistosomiasis control professionals with snail survey experiences of 6 years and less (75%) and more than 6 years (90%) (χ² = 7.792, P < 0.05). The accuracy, sensitivity, specificity and AUC of the AI-enabled intelligent snail identification system were 88%, 100%, 76% and 0.88 for recognition of O. hupensis robertsoni snail sample images, and there was no significant difference in the accuracy of recognition of O. hupensis robertsoni snail sample images between the AI-enabled intelligent snail identification system and artificial identification (χ² = 0.204, P > 0.05). In addition, there was no significant difference in the accuracy of artificial identification for recognition of snail sample images with upward (90%) and downward shell openings (86%) (χ² = 0.379, P > 0.05), and there was a significant difference in the accuracy of artificial identification for recognition of snail sample images between schistosomiasis control professionals with snail survey experiences of 6 years and less and more than 6 years (χ² = 5.604, P_adjusted < 0.025). Conclusions The accuracy of recognition of snail sample images is comparable between the AI-enabled intelligent snail identification system and artificial identification by schistosomiasis control professionals, and the AI-enabled intelligent snail identification system is feasible for recognition of O. hupensis robertsoni and Tricula in Yunnan Province.

Image 1.

Objective To construct a visual intelligent recognition model for Oncomelania hupensis robertsoni in Yunnan Province based on the EfficientNet-B4 model, and to evaluate the impact of data augmentation methods and model hyperparameters on the recognition of O. hupensis robertsoni. Methods A total of 400 O. hupensis robertsoni and 400 Tricula snails were collected from Yongsheng County, Yunnan Province in June 2024, and snail images were captured following identification and classification of 300 O. hupensis robertsoni and 300 Tricula snails. A total of 925 O. hupensis robertsoni images and 1 062 Tricula snail images were collected as a dataset and divided into a training set and a validation set at a ratio of 8:2, while 352 images captured from the remaining 100 O. hupensis robertsoni and 354 images from the remaining 100 Tricula snails served as an external test set. All acquired images were subjected to preprocessing, including cropping and resizing. Three data augmentation approaches were employed, including baseline, Mixup and Gaussian blurring, and model hyperparameters included two optimization algorithms of adaptive moment estimation (Adam) and stochastic gradient descent (SGD), two loss functions of focal loss and cross entropy loss, and two learning rate decay strategies of cosine annealing and multi-step. The intelligent recognition models of O. hupensis robertsoni and Tricula snails were constructed based on the EfficientNet-B4 model, and 7 training strategy groups were generated by combinations of different data augmentation approaches and hyperparameters. The performance of intelligent recognition models was tested with external test sets, and evaluated with accuracy, precision, recall, F1 score, loss, Youden’s index, and the area under the receiver operating characteristic curve (AUC) under different training strategies. Results The variation of loss values was comparable among intelligent recognition models with different data augmentation approaches. The Group 4 model constructed with Mixup and Gaussian blurring data augmentation approaches showed the optimal performance, with an accuracy of 90.38%, precision of 90.07%, F1 score of 89.44%, Youden’s index of 0.81 and AUC of 0.961 in the external test set. The accuracy of models using the SGD optimizer reduced by 29.16% as compared to those using the Adam optimizer (χ² = 81.325, P < 0.001), and the accuracy of models using the cross entropy loss function reduced by 0.80% as compared to the Group 4 model (χ² = 3.147, P > 0.05), while the accuracy of models using the multi-step learning rate decay strategy increased by 0.65% as compared to the Group 4 model (χ² = 0.208, P > 0.05). In addition, the model with the baseline + Mixup + Gaussianblurring data augmentation approach and hyperparameters of Adam optimizer, focal loss function and multi-step learning rate decay strategy showed the highest performance, with an accuracy of 91.03%, precision of 91.97%, recall of 88.11%, F1 score of 90.00%, Youden’s index of 0.82 and AUC values of 0.969 in external test set, respectively. Conclusions The intelligent recognition model of O. hupensis robertsoni based on EfficientNet-B4 model is accurate for identification of O. hupensis robertsoni and Tricula snails in Yunnan Province.

Objective:To compare the characteristics of background mucosa, lesion features, and the efficiency of endoscopic submucosal dissection(ESD)between elderly and non-elderly patients with early gastric cancer(EGC).Methods:This study retrospectively collected data on patients with EGC who underwent ESD treatment at Beijing Hospital from April 2020 to December 2022.The clinical characteristics, background mucosa, lesion features, ESD outcomes, and pathological results of the patients were analyzed to compare the differences between elderly and non-elderly patients.Results:A total of 100 patients with EGC were selected, comprising 57 patients in the elderly group and 43 patients in the non-elderly group, with a total of 111 lesions identified(64 lesions in the elderly group and 47 lesions in the non-elderly group).The proportion of patients with a history of chronic atrophic gastritis was significantly higher in the elderly group(89.5%、51/57)compared to the non-elderly group(74.4%、32/43), with a statistically significant difference( P=0.047).Additionally, the difference in the extent of atrophy between elderly patients with EGC and their non-elderly counterparts was statistically significant( P=0.022).Among these patients, the proportion of those classified as Kimura-Takemoto C0 to C1 in the elderly group(15.6%、10/64)was lower than that in the non-elderly group(40.4%、19/47).In contrast, the proportion of patients classified as C2 to C3 in the elderly group(65.6%、42/64)was higher than that in the non-elderly group(51.1%、24/47), and the proportion of those classified as O1 to O3 in elderly patients(12.5%、8/64)was also higher than in the non-elderly group(4.3%、2/47).Furthermore, the difference in the extent of intestinal metaplasia between elderly and non-elderly patients with early gastric cancer was statistically significant( P=0.007).The overall proportion of total intestinal metaplasia in elderly patients(85.9%、55/64)was significantly higher than that in non-elderly patients(61.7%、29/47).Notably, the proportion of patients exhibiting extensive intestinal metaplasia(intestinal metaplasia present in both the gastric antrum and gastric body)was greater in the elderly group(43.8%、28/64)compared to the non-elderly group(23.4%、11/47).The Kyoto gastric cancer risk endoscopic score for elderly patients with EGC was(2.43±1.28)points, significantly higher than that of the non-elderly group(1.72±1.41)points, with a statistically significant difference observed( t=2.778, P=0.006).No statistically significant differences were observed in the proportions of total resection rates, R0 resections, curative resections, or postoperative complications following ESD when comparing elderly patients with EGC to their non-elderly counterparts. Conclusions:The proportion of extensive atrophy and intestinal metaplasia was higher in the background mucosa of elderly patients with EGC, and correspondingly, the Kyoto endoscopic gastric cancer risk score was elevated.Therefore, endoscopic examinations for elderly patients with chronic atrophic gastritis should be conducted with greater care and comprehensiveness.

Objective:A high performance liquid chromatography (HPLC) method was developed to determine the enzymatic activity of CD38 in blood, which was the major enzyme responsible for consuming nicotinamide adenine dinucleotide (NAD). Additionally, the study aimed to detect the differences in CD38 enzymatic activity among individuals of varying ages and health statuses.Methods:A 50 μl whole blood matrix and enzyme reaction substrate of 150 μl β-NAD at a concentration of 500 μmol/L were selected for the analysis. To eliminate the impact of endogenous β-NAD, the whole blood sample was pre-incubated at 37 ℃ for 20 minutes before adding the substrate. The reaction was terminated by perchloric acid (PCA) after incubation at 37 ℃ for 40 min. The change in product nicotinamide (NAM) before and after the enzymatic reaction was measured by HPLC to calculate the CD38 activity. The linearity, limit of detection, limit of quantification, precision, and stability of the method were evaluated. The CD38 enzymatic activities in 60 healthy volunteers and 30 colorectal cancer patients in blood were determined by the developed method.Results:Pre-incubation at 37 ℃ for 20 minutes eliminated the effect of endogenous β-NAD. The correlation coefficient of NAM was 0.999 in the concentration range of 0.1-3.2 μmol/L, with limit of detection of 0.5 nmol/L and limit of quantification of 2.1 nmol/L. The average within-run imprecision ( CV) and total CV were 3.22%-4.03% and 2.91%-4.70%, respectively. The recovery rate ranged from 94.82% to 96.81%. The CD38 activity of whole blood was stable by storage at 4 ℃ for 48 hours, storage at room temperature for 8 hours, thawing of frozen whole blood at room temperature for 2 hours, or repeated freeze-thawing three times. NAM, NAD standards, and pre-treatment samples were stable after 48 hours at 4 ℃ and 8 hours at room temperature. CD38 activity gradually decreased with increasing concentration of the added CD38 inhibitor 4-aminoquinoline derivative (78c). Measurement of 60 healthy physical examination population samples showed significantly higher CD38 enzyme activity in the elderly group than that in the young group ( t=-2.776, P=0.007) and measurement of 30 colorectal cancer patients showed significantly higher CD38 enzyme activity than that in healthy people ( t=-2.572, P=0.012). Conclusion:The established HPLC method for determining CD38 enzymatic activity is characterized by its simplicity, efficiency, accuracy, and reproducibility. This technique serves as a valuable tool for investigating aging and aging-related diseases.

Objective To investigate the dosimetric differences between volumetric modulated arc therapy (VMAT) with a flattening filter (FF) and flattening filter-free (FFF) VMAT in fractionated stereotactic radiotherapy for brain metastases. Methods Seventeen patients with brain metastases were divided into FF-VMAT group (VMAT plans with the FF mode) and FFF-VMAT group (VMAT plans with the FFF mode). The two groups were compared in terms of target volume dose parameters (D_98%, D_2% and D_mean), the conformal index (CI), the gradient index (GI), the gradient, normal brain tissue dose parameters (V_5Gy, V_10Gy, V_12Gy and D_mean), monitor units, and beam-on time. Results Compared with the FF-VMAT group, the FFF-VMAT group had significantly lower GI (3.33 ± 0.37 vs 3.27 ± 0.35, P = 0.001), a significantly lower gradient [(0.85 ± 0.20) cm vs (0.84 ± 0.19) cm, P = 0.002], a significantly shorter beam-on time [(177.05 ± 62.68) s vs (142.71 ± 34.59) s, P = 0.001], and significantly higher D_2% [(65.69 ± 2.15) Gy vs (66.99 ± 2.03) Gy, P = 0.001] and D_mean [(58.77 ± 1.60) Gy vs (59.95 ± 1.43) Gy, P <0.001]. There were no significant differences in the CI, the D_98% of the target volume, the V_5Gy, V_10Gy, V_12Gy and D_mean of the normal brain tissue, and monitor units between FFF-VMAT and FF-VMAT. Conclusion FFF-VMAT can better protect the normal tissue around the target volume, reduce the beam-on time, and improve treatment efficiency.

Liver transplantation is an effective treatment for the end-stage liver disease. However, hepatic ischemia-reperfusion injury (HIRI) will inevitably occur during liver transplantation, which might lead to early graft dysfunction or aggravate rejection. The underlying protective mechanism remains to be further elucidated. Programmed cell death is an important mechanism of HIRI, and multiple novel types of programmed cell death participate in the pathological process of HIRI. In-depth study of programmed cell death is expected to further improve the therapeutic effect of liver transplantation. In this article, research progresses on apoptosis, autophagy and autophagy-dependent cell death, ferroptosis, necroptosis, pyroptosis, pathanatos and other common programmed cell death patterns in HIRI were reviewed, aiming to provide reference for enhancing the success rate of liver transplantation and improving clinical prognosis of the recipients.

Background: The risk factors affecting the non ⁃ curative resection of endoscopic submucosal dissection (ESD) for early colorectal cancer and precancerous lesions were still lacking large scale studies in China. Aims: To analyze the efficacy of ESD for patients with early colorectal cancer and precancerous lesions, and explore the risk factors affecting the non⁃curative resection. Methods: A total of 229 patients with early colorectal cancer and precancerous lesions underwent ESD from September 2016 to September 2021 at Beijing Hospital were collected. The clinicopathological characteristics, efficacy of ESD were analyzed. Risk factors affecting the non ⁃ curative resection of ESD were analyzed by univariate analysis and Logistic multivariate regression analysis. Results: A total of 255 lesions were found. The en bloc resection rate was 90.2%, R0 resection rate was 87.8%, and curative resection rate was 83.5%. The total incidence of complications was 2.7%. Univariate analysis showed that detection rate of intraluminal protruding tumor (61.0% vs. 24.4%, P<0.05), rate of severe submucosal fibrosis (24.4% vs. 2.4%, P<0.05) were significantly increased in the non⁃curative resection group than in the curative resection group, and positive rate of lifting sign was significantly decreased (80.5% vs. 95.1%, P<0.05). Logistic multivariate regression analysis showed that intraluminal protruding tumor (OR=4.087, 95% CI: 1.523⁃10.968, P= 0.005) and severe submucosal fibrosis (OR=9.609, 95% CI: 1.107⁃83.439, P=0.04) were risk factors for ESD non⁃curative resection of early colorectal cancer and precancerous lesions. Conclusions: R0 resection rate and curative resection rate of ESD for early colorectal cancer and precancerous lesions are high, the incidence of surgical complications is low. Intraluminal protruding tumor, severe submucosal fibrosis are the independent risk factors of non⁃curative resection. Strict evaluation should be carried out before operation and reasonable treatment strategies should be formulated accordingly.

Objective Aiming at the clinical problem of the low matching degree with the patient’s anatomical morphology for traditional cervical fusion cage, a cervical fusion cage with the function of adjustable height and the shape matched with the vertebral body was established, and its biomechanical properties were evaluated. Methods A cervical C4-5 segment fusion model was established according to anterior cervical discectomy and fusion (ACDF), so as to simulate different motion conditions, i.e. anterior flexion, posterior extension, left/right lateral flexion, left/right rotation, and stress of the fusion cage and vertebral endplate was calculated. After three-dimensional (3D) printing of the fusion cage, an in vitro mechanical experiment was conducted to explore safety and stability of the fusion cage. ResultsThe fusion cage could keep the range of motion (ROM) of cervical vertebrae at the fusion segment with 1°-2.8° and reduce the ROM to 40%-80% of the natural segment. In the in vitro compression test, the yield load of the fusion cage was (2 721.67±209) N, which met the maximum demand of the physiological load in service state. Conclusions The designed fusion device with adjustable height shows better biomechanical properties and can reduce the selection step in operation.

Objective:To investigate the feasibility of predicting lateral cervical lymph node metastasis (LLNM) in patients with papillary thyroid carcinoma (PTC) based on the nomogram constructed by dual-energy CT data.Methods:In total 417 patients with PTC confirmed by pathology in Tianjin First Central Hospital from January 2015 to December 2018 were retrospectively analyzed as a training group. Internal validation was conducted, including 139 patients in the LLNM group and 278 patients in the non-LLNM group. A total of 169 PTC patients from January 2019 to June 2020 were included as an external validation group, including 58 patients in the LLNM group and 111 patients in the non-LLNM group. The morphological characteristics of the primary thyroid lesions on dual-energy CT iodine maps were analyzed, including tumor location, maximum diameter, calcification, and extrathyroidal extension (ETE). Iodine concentration (IC) of the PTC parenchyma and the internal carotid artery on the same level in the arterial and venous phases were measured, and normalized iodine concentration (NIC) was calculated. The independent risk factors for LLNM were obtained by univariate and multivariate logistic regression analysis. Base on the results, a prediction model was constructed and expressed in the form of a nomogram. The internal and external validation of the model was carried out using ROC curve.Results:Multivariate binary logistic regression analysis showed that the lesion location in the upper polar of the thyroid, the presence of ETE, IC in arterial phase>2.9 mg/ml, IC in the venous phase>3.2 mg/ml, and NIC in the arterial phase>0.21 were independent risk factors for LLNM prediction. The nomogram based on the above factors was constructed with an area under the ROC curve (AUC) of 0.895 (95%CI 0.862-0.923). With a cut-off value of 0.79, the sensitivity and specificity were 86.3% and 75.2%, respectively. As for the external validation group, the AUC of the model was 0.887 (95%CI 0.830-0.931), with the sensitivity of 82.8%, and the specificity of 81.1%.Conclusion:The application values of the nomogram model based on dual-energy CT data in preoperative evaluation of the possibility of LLNM of PTC patients has been verified. The model constructed in this study might be helpful with the individualized treatment in a certain degree.