ObjectiveThe widespread adoption of portable fundus cameras for primary care and community screening is hindered by limitations in current autofocus(AF) technologies. Image-based methods relying on sharpness evaluation require iterative searches, resulting in slow convergence, while projection-based techniques are susceptible to optical artifacts and calibration errors. To address these challenges, this study introduces a novel AF system based on direct wavefront sensing, designed to deliver simultaneous high speed, high precision, and operational robustness within the compact form factor essential for portable ophthalmic devices. MethodsOur approach fundamentally reimagines the AF process by directly measuring the ocular wavefront aberration. We developed a custom portable fundus camera integrating a miniaturized Shack-Hartmann wavefront sensor (SHWS) into the optical path. An 850 nm laser diode projects a point source onto the retina via oblique illumination to minimize corneal reflections. Light scattered from this spot carries the eye’s refractive error through the imaging optics and is directed to the SHWS, positioned at a plane optically conjugate to the primary color CMOS imaging sensor. A microlens array within the SHWS samples the incident wavefront, generating a pattern of focal spots on a CCD. Real-time centroid analysis of these spots provides a map of local wavefront slopes. These measurements are processed through a singular value decomposition (SVD) algorithm to fit a Zernike polynomial basis set, enabling real-time reconstruction of the wavefront phase. The defocus component (S) is extracted from the second-order Zernike coefficients, providing a direct, quantitative measure of the refractive error in diopters. This value serves as a precise error signal in a closed-loop control system, which commands a voice-coil actuated focusing lens to its null position in a single, deterministic step, eliminating the need for iterative search algorithms. ResultsComprehensive evaluation demonstrated the system’s high performance. Testing on a calibrated model eye (OEMI-7) established a highly linear relationship between the computed defocus S and the focusing lens position across a ±20 Diopter (D) compensation range, achievable within a 5 mm mechanical travel. The system achieved a focusing precision of 0.08 D, corresponding to an 18-fold improvement over a conventional projection spot-size method tested under identical conditions. The total focus acquisition time, encompassing wavefront measurement, computation, and lens actuation, averaged under 0.5 s. Clinical validation with 25 human volunteers (50 eyes, refractive range -15 D to +10 D) confirmed practical efficacy. The wavefront-sensing AF succeeded in 92% of attempts with a mean time of 0.5 s, substantially outperforming a projection-based benchmark which achieved only a 32% success rate with an average time of 4.25 s. The system provided instantaneous directional guidance and maintained stability during minor ocular movements. Objective assessment of image quality, via amplitude contrast of retinal vasculature, showed consistent and significant enhancement following AF correction across the entire tested diopter range. ConclusionThis work successfully implements and validates a direct wavefront-sensing autofocus paradigm for portable fundus cameras. By directly quantifying and compensating for the optical defocus aberration, this method bypasses the fundamental limitations of image-processing and projection-based techniques, enabling rapid, precise, and deterministic diopter compensation. The developed system delivers an exceptional combination of a wide operational range (±20 D), high accuracy (0.08 D), fast convergence (0.5 s), and a compact physical footprint. This technology provides a practical and high-performance focusing solution capable of enhancing the reliability, throughput, and diagnostic utility of portable retinal imaging in large-scale screening applications. Future efforts will be directed towards system cost optimization and performance adaptation for diverse ocular conditions.

ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

Objectives: . Idiopathic sudden sensorineural hearing loss (ISSNHL) is a significant cause of hearing loss. Intratympanic steroid injection (ITSI) is commonly used as an initial or salvage treatment; however, the lack of a standardized treatment protocol has resulted in variability in clinical practice. In addition, no efficient prediction model currently exists to support personalized management. Therefore, this study aimed to develop tailored management strategies for ISSNHL using a machine-learning model. Methods: . This retrospective multicenter cohort study was conducted between January 2015 and December 2020, with data analysis performed between January 2021 and March 2024. Patients were selected based on the International Classification of Diseases, 10th Revision criteria for ISSNHL, along with relevant medication and procedure codes. Patients with pure-tone audiogram results not meeting ISSNHL criteria, better initial hearing in the affected ear, an identifiable etiology, no post-treatment audiogram, or delayed treatment (>6 weeks) were excluded. We included 770 patients diagnosed with ISSNHL who received ITSI. The primary outcome was the area under the receiver operating characteristic curve for prediction performance. Recovery status was determined using the last pure-tone audiogram. Modeling was conducted on the Quanta for Medical Care AI platform using five machine-learning algorithms and a nested cross-validation framework, in which feature selection and hyperparameter tuning were performed in the inner folds and model performance was evaluated in the outer folds. Results: . A random forest classifier outperformed the other models in predicting hearing outcomes, achieving an area under the receiver operating characteristic curve of 0.788. Time to ITSI was the most influential treatment-related factor, with ITSI administered within 10 days of hearing loss being associated with better outcomes. This model can be used to provide personalized prognostic estimates under different treatment protocols. Conclusion . The machine-learning-based prediction model facilitates personalized treatment strategies and timely treatment adjustments for ISSNHL, thereby optimizing the likelihood of complete recovery.

Objective:To investigate the clinical characteristics of patients with multiple myeloma(MM)complicated by bone lesions and the risk factors associated with bone lesions.Methods:The clinical data of 294 newly diagnosed MM patients in Gansu Provincial Hospital from January 2017 to June 2021 were retrospectively analyzed.The patients were divided into the bone lesion group(154 cases)and the non-bone lesions group(140 cases)based on the presence of absence of bone lesions at diagnosis.The general data and laboratory parameters were compared between the two groups.The risk factors for bone lesions in MM patients were analyzed by logistic regression analysis,and the characteristic(ROC)curves were plotted to assess the predictive value of each risk factor for the occurrence of bone lesions in MM patients.Results:Compared to the non-bone lesion group,the bone lesion group had significantly higher serum calcium levels and significantly greater proportions of patients with Durie-Salmon(DS)stage Ⅲ,and bone pain(all P＜0.05).Logistic regression analysis showed that elevated serum calcium(OR=5.135,95％CI:1.931-13.653,P=0.001),DS stage Ⅲ(OR=1.841,95％CI:1.019-3.328,P=0.043),and bone pain(OR=8.208,95％CI:4.761-14.151,P＜0.001)were independent risk factors for bone lesions in MM patients.ROC curve analysis showed that serum calcium(AUC=0.619,95％CI:0.555-0.683,P＜0.001)and bone pain(AUC=0.743,95％CI:0.692-0.793,P＜0.001)had predictive value for bone lesions in MM patients.Conclusion:MM patients have a high incidence of bone lesions,and active monitoring and management of risk factors may improve treatment outcomes and prognosis.

The purpose of this study was to evaluate the application value of targeted next-generation sequencing(tNGS)for rapid detection of Mycobacterium tuberculosis.From September 2021 to October 2024,1 178 clinical samples from hospitalized pa-tients at the Non-tuberculous Mycobacteria Disease Diagnosis and Treatment Center of Guangzhou Chest Hospital were collected,in-cluding 272 sputum samples,871 broncho alveolar lavage fluid(BALF)samples,and 35 other samples.These samples were analyzed with tNGS and the liquid culture+DNA microarray chip method(referred to as the"culture identification method"),and the detection results between methods were compared.The numbers of Mycobacterium tuberculosis complex(MTBC)samples identified with tNGS and the culture identification method were 172 and 127,respectively,and the numbers of non-tuberculous mycobacteria(NTM)samples detected were 517 and 474,respectively.The detection rate of Mycobacterium was 56.88%with tNGS and was significantly lower(49.49%)with the culture identification method(χ2=13.27,P<0.05).For NTM identification,when sputum was used,the detection rate of tNGS was higher than that of the culture identification method(χ2=24.14,P<0.05).However,when BALF was used,no significant difference in detection rates was observed between tNGS and the culture identification method(χ2=3.97,P=0.06).When different sample types from the same patient were analyzed with tNGS for NTM,BALF was found to be a better choice than sputum(χ2=4.05,P<0.05).However,with the culture identification method,we observed no significant difference between sample types(χ2=2.72,P=0.10).Furthermore,both the MTBC and NTM sequences detected with tNGS were significantly higher in the culture-positive group than the culture-negative group.With increasing sequence numbers,the proportion of cases in the culture-positive group progressively rose,whereas an inverse trend was observed in the culture-negative group.With clinical diagno-sis as the reference standard,the sensitivity,specificity,and Kappa value of tNGS and the culture identification method in identifying mycobacterial diseases were 85.41%,91.32%,0.74 and 74.73%,93.15%,0.63,respectively.Compared with the culture identifica-tion method,tNGS for rapid detection of Mycobacterium exhibited excellent sensitivity,specificity and consistency,and its timeliness should help clinicians make precise individualized treatment plans earlier.

Objective:To observe the clinical effect of Cangfu Daotan Decoction combined with ethinylestradiol and cyproterone acetate,metformin in patients with spleen deficiency phlegm-dampness type polycystic ovary syndrome(PCOS)accompanied by insulin resistance.Methods:80 patients with PCOS accompanied by insulin resistance who were admitted to our hospital from June 2023 to June 2024 were included.The patients were divided into control group(treated with ethinylestradiol and cyproterone acetate,metformin)and study group(treated with modified Cangfu Daotan Decoction combine with ethinylestradiol and cyproterone acetate,metformin)by the random number table method,with 40 cases in each group.The total clinical effective rate,TCM syndrome score,sex hormone levels[prolactin(PRL),estradiol(E2),follicle-stimulating hormone(FSH),testosterone(T),and luteinizing hormone(LH)],insulin resistance indicators[fasting insulin(FINS),fasting blood glucose(FBG),homeostatic model assessment of insulin resistance(HOMA-IR)],and endometrial receptivity[pulsatile index(PI),resistance index(RI),endometrial thickness(ET),ovulation rate]and the occurrence of adverse reactions were compared between the two groups.Results:Compared with the control group after treatment,the total clinical effective rate,PRL,E2,FSH level,ET and the ovulation ratein the study group were higher,while the main symptoms,secondary symptoms,total score,T,LH levels,FBG,FINS levels,HOMA-IR,PI and RI were lower(P＜0.05).There was no difference in the incidence of adverse reactions between the two groups(P＞0.05).Conclusion:Cangfu Daotan Decoction combined with ethinylestradiol and cyproterone acetate,metformin in patients with spleen deficiency phlegm-dampness type PCOS accompanied by insulin resistance,can improve the clinical symptoms,increase the total clinical effective rate,it may be related to the regulation of insulin resistance,sex hormone levels,and endometrial receptivity,it is safe and reliable.

Objective To elucidate the impact of chronic kidney disease(CKD)on the clinical outcomes of patients with left main coronary artery disease(LMCAD)undergoing intravascular ultrasound(IVUS)-guided percutaneous coronary intervention(PCI).Methods This retrospective study enrolled consecutive patients with LMCAD who underwent IVUS-guided PCI at Wuhan Asia Heart Hospital between January 2017 and December 2020.Patients were stratified into CKD and non-CKD groups according to the presence of CKD.Clinical data were systematically retrieved from the electronic health record system.Demographic,clinical,and angiographic characteristics were compared between groups.The primary endpoint was major adverse cardiovascular events(MACE),defined as a composite of all-cause mortality,myocardial infarction,and ischemic stroke.Results A total of 325 LMCAD patients[mean age(62.56±9.86)years;73.54％male]were included,with 31 patients(9.54％)in the CKD group.During a median follow-up of 5 years,CKD patients exhibited significantly older age[(70.13±9.77)years vs.(61.77±9.54)years,P＜0.001],higher prevalence of three-vessel disease(64.52％vs.38.10％;P=0.040)and left main bifurcation lesion(45.16％vs.37.76％,P=0.011),greater IVUS-detected calcification burden(P=0.029),and higher median SYNTAXⅡ scores[(34.10(30.30,39.25)vs.26.75(22.42,31.58),P＜0.001)].The cumulative incidence of MACE was significantly higher in the CKD group compared to the non-CKD group(32.26％vs.9.18％,P＜0.001).Univariate Cox regression analysis and Kaplan-Meier survival curves confirmed a 5.877-fold increased risk of MACE in CKD patients(95％CI 2.765-12.494).After adjusting for age and cardiac function,CKD remained an independent predictor of MACE(HR 3.611,95％CI 1.634-7.978).Conclusions LMCAD patients with concomitant CKD present with advanced age,impaired cardiac function,more extensive coronary disease,and severe calcification.The presence of CKD is associated with a significantly worse long-term prognosis.

Objective To explore the effects of different education and examination methods on the examination results during the screening/evaluation of patent foramen ovale by contrast-enhanced transcranial Doppler(cTCD).Methods Patients who underwent cTCD screening/evaluation for patent foramen ovale in our hospital from May 2023 to February 2024 were retrospectively selected as the research subjects.The patients were divided into the observation group and the control group according to different education and examination methods during the peri-examination period.Patients who received video education,modified Valsalva maneuver,and injection of contrast agent with 20 mL syringe were included into the observation group,and patients who received artificial education,Valsalva maneuver,and injection of contrast agent with 10 mL syringe were included into the control group.The positive detection rate of patent foramen ovale,right-to-left shunt microbubble grading during Valsalva/modified Valsalva maneuver,systolic blood flow velocity,pulsatility index(PI),resistive index(RI),examination duration,total physician-patient communication time,whether occlusion surgery was performed,and patient satisfaction were compared between the two groups.Results The positive detection rate of patent foramen ovale by cTCD(82.93%vs.95.92%),the detection rate of the maximum amout(grade Ⅲ)of microbubbles(39.02%vs.61.22%),the total physician-patient communication time during the peri-examination period[11.30(10.00,14.00)minutes vs.8.23(7.00,10.00)minutes],the rate of occlusion surgery(48.78%vs.73.47%),and the total patient satisfaction(80.49%vs.91.84%)showed statistically significant differences between the control group and the observation group(P<0.05).Additionally,the receiver operating characteristic(ROC)curve analysis showed that the area under the curve(AUC)for diagnosing patent foramen ovale were 0.718 in the control group and 0.855 in the observation group.Conclusion Peri-examination interventions such as video education,modified Valsalva maneuver,and injection of contrast agent with 20 mL syringe can improve the positive detection rate of patent foramen ovale,reduce ineffective physician-patient communication,and improve patient satisfaction.