Testicular histology based on testicular biopsy is an important factor for determining appropriate testicular sperm extraction surgery and predicting sperm retrieval outcomes in patients with azoospermia. Therefore, we developed a deep learning (DL) model to establish the associations between testicular grayscale ultrasound images and testicular histology. We retrospectively included two-dimensional testicular grayscale ultrasound from patients with azoospermia (353 men with 4357 images between July 2017 and December 2021 in The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China) to develop a DL model. We obtained testicular histology during conventional testicular sperm extraction. Our DL model was trained based on ultrasound images or fusion data (ultrasound images fused with the corresponding testicular volume) to distinguish spermatozoa presence in pathology (SPP) and spermatozoa absence in pathology (SAP) and to classify maturation arrest (MA) and Sertoli cell-only syndrome (SCOS) in patients with SAP. Areas under the receiver operating characteristic curve (AUCs), accuracy, sensitivity, and specificity were used to analyze model performance. DL based on images achieved an AUC of 0.922 (95% confidence interval [CI]: 0.908-0.935), a sensitivity of 80.9%, a specificity of 84.6%, and an accuracy of 83.5% in predicting SPP (including normal spermatogenesis and hypospermatogenesis) and SAP (including MA and SCOS). In the identification of SCOS and MA, DL on fusion data yielded better diagnostic performance with an AUC of 0.979 (95% CI: 0.969-0.989), a sensitivity of 89.7%, a specificity of 97.1%, and an accuracy of 92.1%. Our study provides a noninvasive method to predict testicular histology for patients with azoospermia, which would avoid unnecessary testicular biopsy.
Humans ; Male ; Azoospermia/diagnostic imaging* ; Deep Learning ; Testis/pathology* ; Retrospective Studies ; Adult ; Ultrasonography/methods* ; Sperm Retrieval ; Sertoli Cell-Only Syndrome/diagnostic imaging*

The choice of biopsy method is critical in diagnosing prostate cancer (PCa). This retrospective cohort study compared systematic biopsy (SB) or cognitive fusion-targeted biopsy combined with SB (CB) in detecting PCa and clinically significant prostate cancer (csPCa). Data from 2572 men who underwent either SB or CB in Fudan University Shanghai Cancer Center (Shanghai, China) between January 2019 and December 2023 were analyzed. Propensity score matching (PSM) was used to balance baseline characteristics, and detection rates were compared before and after PSM. Subgroup analyses based on prostate-specific antigen (PSA) levels and Prostate Imaging-Reporting and Data System (PI-RADS) scores were performed. Primary and secondary outcomes were the detection rates of PCa and csPCa, respectively. Of 2572 men, 1778 were included in the PSM analysis. Before PSM, CB had higher detection rates for both PCa (62.9% vs 52.4%, odds ratio [OR]: 1.54, P < 0.001) and csPCa (54.9% vs 43.3%, OR: 1.60, P < 0.001) compared to SB. After PSM, CB remained superior in detecting PCa (63.1% vs 47.9%, OR: 1.86, P < 0.001) and csPCa (55.0% vs 38.2%, OR: 1.98, P < 0.001). In patients with PSA 4-12 ng ml -1 (>4 ng ml -1 and ≤12 ng ml -1 , which is also applicable to the following text), CB detected more PCa (59.8% vs 40.7%, OR: 2.17, P < 0.001) and csPCa (48.1% vs 27.7%, OR: 2.42, P < 0.001). CB also showed superior csPCa detection in those with PI-RADS 3 lesions (32.1% vs 18.0%, OR: 2.15, P = 0.038). Overall, CB significantly improves PCa and csPCa detection, especially in patients with PSA 4-12 ng ml -1 or PI-RADS 3 lesions.
Humans ; Male ; Prostatic Neoplasms/diagnosis* ; Propensity Score ; Retrospective Studies ; Middle Aged ; Aged ; Image-Guided Biopsy/methods* ; Prostate-Specific Antigen/blood* ; Prostate/diagnostic imaging*

Objective To explore a deep learning-based automatic bone age estimation model for elbow joint X-ray images of Chinese Han adolescents and children and evaluate its performance.Methods A total of 943(517 males and 426 females)elbow joint frontal view X-ray images of Chinese Han ado-lescents and children aged 6.00 to<16.00 years were collected from East,South,Central and North-west China.Three experimental schemes were adopted for bone age estimation.Scheme 1:Directly in-put preprocessed images into the regression model;Scheme 2:Train a segmentation network using"key elbow joint bone annotations"as labels,then input segmented images into the regression model;Scheme 3:Train a segmentation network using"full elbow joint bone annotations"as labels,then in-put segmented images into the regression model.For segmentation,the optimal model was selected from U-Net,UNet++and TransUNet.For regression,VGG16,VGG19,InceptionV2,InceptionV3,ResNet34,ResNet50,ResNet101 and DenseNet121 models were selected for bone age estimation.The dataset was randomly split into 80%(754 samples)for training and validation for model fitting and hyperparameter tuning,and 20%(189 samples)as an internal test set to test the performance of the trained model.An additional 104 elbow joint X-ray images from the same demographic and age group were col-lected and used as an external test set.Model performance was evaluated by comparing the mean ab-solute error(MAE),root mean square error(RMSE),accuracies within±0.7 years(P±0.7 years)and±1.0 years(P±1.0 years)between the estimated age and the actual age,and by drawing radar charts,scat-ter plots,and heatmaps.Results When segmented with Scheme 3,the UNet++model achieved good segmentation performance with a segmentation loss of 0.000 4 and an accuracy of 93.8%at a learning rate of 0.000 1.In the internal test set,the DenseNet121 model with Scheme 3 yielded the best results with MAE,P±0.7 years and P±1.0 years being 0.83 years,70.03%,and 84.30%,respectively.In the external test set,the DenseNet121 model with Scheme 3 also performed best,with an average MAE of 0.89 years and an average RMSE of 1.00 years.Conclusion When performing automatic bone age estima-tion using elbow joint X-ray images in Chinese Han adolescents and children,it is recommended to use the UNet++model for segmentation.The DenseNet121 model with Scheme 3 achieves optimal per-formance.Using segmentation networks,especially that trained with annotation areas encompassing the full elbow joint including the distal humerus,proximal radius,and proximal ulna,can improve the ac-curacy of bone age estimation based on elbow joint X-ray images.

Objective To explore a deep learning network model suitable for bone age estimation using shoulder joint X-ray images in Chinese Han adolescents.Methods A retrospective collection of 1 286 shoulder joint X-ray images of Chinese Han adolescents aged 12.0 to<18.0 years(708 males and 578 females)was conducted.Using random sampling,approximately 80%of the samples(1 032 cases)were selected as the training and validation sets for model learning,selection and optimization,and the other 20%samples(254 cases)were used as the test set to evaluate the model's generalization ability.The original single-channel shoulder joint X-ray images and dual-channel inputs combining original images with segmentation labels(manually annotated shoulder joint regions multiplied pixel-by-pixel with original images,followed by segmentation via the U-Net++network to retain only key shoulder joint region information)were respectively input into four network models,namely VGG16,ResNet18,ResNet50 and DenseNet121 for bone age estimation.Additionally,manual bone age estimation was con-ducted on the test set data,and the results were compared with the four network models.The mean absolute error(MAE),root mean square error(RMSE),coefficient of determination(R2),and Pear-son correlation coefficient(PCC)were used as main evaluation indicators.Results In the test set,the bone age estimation results of the four models with dual-channel input of shoulder joint X-ray images outperformed those with single-channel input in all four evaluation indicators.Among them,DenseNet121 with dual-channel input achieved best results with MAE of 0.54 years,RMSE of 0.82 years,R2 of 0.76,and PCC(r)of 0.88.Manual estimation yielded an MAE of 0.82 years,ranking second only to dual-channel DenseNet121.Conclusion The DenseNet121 model with dual-channel input combined with original images and segmentation labels is superior to manual evaluation results,and can effectively estimate the bone age of Chinese Han adolescents.

Background Maternal atmospheric pollution during pregnancy may alter fetal intrauterine development programming, thereby increasing the risk of childhood obesity in the future. Objective To investigate the effects of atmospheric pollution exposure during pregnancy on the incidence of childhood obesity in offspring. Methods English databases (PubMed, Web of Science, and Medline) and Chinese databases (Wanfang Data Knowledge Service Platform, China National Knowledge Infrastructure, and VIP Information Chinese Journal Service Platform) were searched for literature reporting exposure to atmospheric pollution during pregnancy and childhood obesity published from 1 January 2000 to 31 August 2023. The quality of the included literature was evaluated using the quality assessment tools for observational cohort and cross-sectional studies recommended by the US National Institutes of Health. Results Twenty-four studies meeting the inclusion criteria were identified and the associated atmospheric pollutants included particulate matter, ozone, nitrogen oxide, carbon oxide, and sulfur oxide. In comparison to the non-exposed group, prenatal exposure to various common atmospheric pollutants were significantly associated with an elevated risk of childhood obesity in offspring. Conclusion Maternal exposure to atmospheric pollution during pregnancy is associated with an elevated risk of childhood obesity in subsequent years. Future studies should pay more attention to the effects of atmospheric pollution on the distribution of children's body fat and metabolic development, and further identify potential mechanisms of atmospheric pollutant exposure leading to childhood obesity.

Objective:To assess the long-term outcome of sequential radical surgery after immune combined with targeted therapy for patients with initially unresectable hepatocellular carcinoma (HCC).Methods:Clinical data of 100 patients with initially unresectable HCC undergoing sequential radical surgery after immune combined with targeted therapy at the Faculty of Hepato-Pancreato-Biliary Surgery of Chinese PLA General Hospital from December 2018 to August 2023 were prospectively collected, including 87 males and 13 females, with a median age of 55 (24-73) years. The pre-treatment tumor staging was determined using the China liver cancer staging (CNLC). The efficacy of immune combined with targeted therapy was accessed using the modified response evaluation criteria in solid tumor (mRECIST). The cycles of immune combined with targeted therapy were analyzed. The tumor residual of resected tissue was analyzed through a standard pathological protocol. The prognosis was analyzed using the Kaplan-Meier method.Results:Upon initial diagnosis, there were 46 cases (46.0%) staged CNLC-Ⅲa and 40 (40.0%) staged CNLC-Ⅲb. There were also 14 cases (14.0%) staged CNLC-Ⅰb, Ⅱa, and Ⅱb who underwent immune combined with targeted therapy due to rupture of tumor or insufficient liver remnant. All patients received a median of 5 (3-28) cycles of immune combined with targeted therapy and underwent radical surgery after successful conversion. According to mRECIST, 14 (14.0%) were determined as complete remission, 63 (63.0%) as partial remission, 18 (18.0%) as stable disease, and 5 (5.0%) as disease progression. Of 24 (24.0%) were defined as pathologically complete remission by postoperative pathology. Furthermore, pathological tumor residue was less than 10% in 61 (61.0%) cases and less than 50% in 82 (82.0%) cases. The 1, 3, and 5 year-overall survival rates of patients were 98.0%, 83.1%, and 74.5%, respectively. The 1, 2 and 3 year-recurrence-free survival rates were 67.5%, 54.8%, and 49.6%, respectively.Conclusion:Sequential radical surgery after immune combined with targeted therapy benefits the long-term survival of patients with initially unresectable HCC.

Objective:To assess the diagnostic performance of 68Ga-PSMA PET/CT and mpMRI in initial staging of prostate cancer. Methods:A retrospective analysis was conducted on patients with initial diagnosis of prostate cancer who underwent prostatectomy at Fudan University Shanghai Cancer Center from December 2021 to June 2023. All had biopsy-confirmed prostate cancer preoperatively and had not undergone any anti-tumor treatment. Prior to surgery, all patients underwent 68Ga-PSMA PET/CT and mpMRI scans. The surgical samples were processed by whole-mount slides pathology. Thirty patients were included, with a median age of 68 years (range 67-76 years). The preoperative median PSA level was 17.91 ng/ml (range 9.41-39.53 ng/ml). The median interval between the two examinations was 21.00 days (range 2.75-35.50 days). Based on the postoperative whole-mount slides pathology which was the gold standard, we compared the sensitivity and specificity of 68Ga-PSMA PET/CT and mpMRI in diagnosing extraprostatic extension, seminal vesicle invasion, and intraprostatic tumor lesion. Results:According to the postoperative pathology, among the 30 patients, 12 had extraprostatic extension (EPE), 8 had seminal vesicle invasion (SVI). Among the 50 lesions found before surgery, postoperative pathology showed that 38 of them were clinically significant prostate cancer lesions (76.0%, 38/50). Using whole-mount pathology, preoperative mpMRI, and 68Ga-PSMA PET/CT imaging for triple fusion comparison, the sensitivity and specificity of 68Ga-PSMA PET/CT in diagnosing EPE were 66.7% (8/12) and 88.9% (16/18), respectively. The sensitivity and specificity of mpMRI in diagnosing EPE were 50.0% (6/12) and 83.3% (15/18), respectively. In diagnosing SVI, the sensitivity and specificity of 68Ga-PSMA PET/CT were 50.0% (4/8) and 86.3% (19/22), respectively. The sensitivity and specificity of mpMRI were 75.0% (6/8) and 90.9% (20/22), respectively. When diagnosing clinically significant tumor lesions within the prostate, the sensitivity of 68Ga-PSMA PET/CT was 89.5% (34/38), which was significantly better than the mpMRI of 63.2% (24/38), with no significant difference in specificity. Conclusions:The sensitivity of 68Ga-PSMA PET/CT in diagnosing intraprostatic lesions was significantly higher than that of mpMRI, with no significant difference in specificity.

Objective To design a wearable cardiopulmonary monitoring system and validate its performance through preliminary human trials.Methods The wearable cardiopulmonary monitoring system was composed of a data collector,a wearing vest and an information management platform.The data collector used an EFM32GG330 SCM as the main microcon-troller unit(MCU),which included a respiratory modulation module,an ECG modulation module,a body position modulation module,a wireless communication module(involving in a Bluetooth module and a Wi-Fi module),a storage module and a power management module.The wearable vest had a cardigan-type structure,and was equipped with ECG sensors and respiratory motion sensors at its inner side.The information management platform was developed with Client/Server(C/S)architecture and Java/JavaScript.The system developed was compared with Mindray's IPM10 Patient Monitor routinely used in hospitals through preliminary human trials to verify its effectiveness in monitoring human heart rate and respiratory rate.Results The system developed could continuously monitor the human heart rate and respiratory rate for a long time,and the monitoring results had high consistency with those of Mindray's IPM10 Patient Monitor.Conclusion The system can be used for medical monitoring of cardiopulmonary indicators during training or exercise,providing accurate physiological information for health management.[Chinese Medical Equipment Journal,2024,45(4):51-55]

The problems caused by proximal contact loss(PCL)of dental implants have been a mainstream research topic in recent years,and scholars are unanimously committed to analyzing their causes and related factors,aiming to identify solutions to the problems related to PCL.The effects of the anterior component of force(ACF),the lifelong re-molding of the adult craniofacial jaw and alveolar socket,and the osseointegration characteristics of dental implants are the main causes of PCL.On the one hand,the closing movement of the mandible causes the ACF of the tooth to move through the posterior molar cusp.Moreover,drifting between the upper and lower posterior teeth and mandibular anteri-or teeth can cause the anterior teeth of the upper and lower jaws to be displaced labially.On the other hand,reconstruc-tion of the jaw,alveolar socket and tooth root,the forward horizontal force of the masticatory muscles,the dynamic com-ponent of the jaw and the forward force generated by the oblique plane of the tooth cusp can cause the natural tooth to experience near-middle drift.Additionally,natural teeth can shift horizontally and vertically and rotate to accommodate remodeling of the stomatognathic system and maintain oral function.Nevertheless,the lack of a natural periodontal mem-brane during implant osseointegration,the lack of a physiological basis for near-medium drift,the small average degree of vertical motion and the integrated silence of dental implants without the overall drift characteristics of natural teeth increases the probability of PCL.The high incidence of PCL is clearly associated with the duration of prosthesis delivery and the mesial position;but it is also affected by the magnitude of the bite force,occlusion,the adjacent teeth,restora-tion design,implant location,jaw,and patient age and sex.PCL has shown a significant correlation with food impaction,but not a one-to-one correspondence,and did not meet the necessary and sufficient conditions.PCL is also associated with peri-implant lesions as well as dental caries.PCL prevention included informed consent,regular examinations,se-lection of retention options,point of contact enhancement,occlusal splints,and the application of multipurpose digital crowns.Management of the PCL includes adjacent contact point additions,orthodontic traction,and occlusal adjust-ment.Existing methods can solve the problem of food impaction in the short term with comprehensive intervention to seek stable,long-term effects.Symmetric and balanced considerations will expand the treatment of issues caused by PCL.

Objective To explore the characteristics of speech prosody chunking,affect,focus and interaction function in patients with non-fluent aphasia after stroke. Methods From December,2023 to March,2024,patients with non-fluent aphasia after stroke(experimental group,n=25)and healthy subjects(control group,n=25)were recruited from Shanghai Fifth Rehabilitation Hospital,and the two groups were compared in terms of chunking,affect,focus and interaction performance using a self-ad-ministered speech prosody function assessment tool and acoustic analyzing instrument. Results In chunking prosody,there were differences in auditory discrimination(Z=-4.856,P<0.001),auditory recog-nition of"two chunks"(Z=-3.024,P=0.002),and subjective scores,total pause duration,average pause dura-tion,average syllabic duration and structural sonic rate(|Z|>2.911,P<0.001)of imitating"two chunks"and au-tonomously outputing"two chunks"and"three chunks"between two groups.In affect prosody,there were differ-ences in listening to distinguish different emotions(Z=-3.322,P=0.001),listening to identify"like emotion"(Z=-3.481,P=0.001),and the subjective score(Z=-6.214,P<0.001),average intensity(Z=-3.581,P<0.001)and standard deviation of intensity(Z=-3.181,P=0.001)of imitating"like emotion"between two groups;there were differences in the subjective score(Z=-5.510,P<0.001),average intensity(Z=-3.429,P=0.001),standard deviation of intensity(Z=-4.777,P<0.001)and fundamental frequency variation range(t=1.852,P=0.029)of autonomously outputing"like emotion"between two groups.In focus prosody,there were differences in listening to distinguish different focuses(Z=-2.658,P=0.008),listening to identify the"end of sentence"focus(Z=-2.023,P=0.011),and the peak stress intensity of imitating the"first sentence"focus(t=8.294,P<0.001);the subjective score(|Z|>5.102,P<0.001)and peak stress intensity(t>8.298,P<0.001)of imitating the"mid-sentence"and"end-sentence",and the subjective score(Z=-4.371,P<0.001)and peak stress intensity(t=8.842,P<0.001)of autonomously outputing the focus of the"first sentence"between two groups;there were differences in the subjective score(|Z|>4.970,P<0.001),peak stress intensity(t>9.373,P<0.001)and stress duration(|t|>2.912,P<0.01)of autonomously outputing the focus of the"mid-sentence"and"end-sentence".In the interaction prosody,there were no significant difference in auditory resolution and au-ditory recognition between two groups(P>0.05),however,there were significant differences in the subjective score(|Z|>4.938,P<0.001)and the fundamental frequency slope k(|t|>4.609,P<0.001)of the interrogative tone both in imitative output and autonomous output. Conclusion Patients with non-fluent aphasia showed significant abnormalities in the recognition and expression of"2 chunks""like emotion""sentence-final focus"and"questioning tone."