To develop a spatiotemporal hybrid perception deep learning (DL) model using contrast-enhanced ultrasound (CEUS) videos for diagnosis of hepatocellular carcinoma (HCC) in liver imaging reporting and data system (LI-RADS) category LR-M liver nodules.

This study conducted a retrospective analysis of patients diagnosed with LR-M nodules by CEUS from the First Affiliated Hospital of Sun Yat-sen University and the First Affiliated Hospital of Guangxi Medical University. Using pathological and comprehensive diagnostic results as the gold standard, the patients were categorized into HCC and non-HCC. A multi-scale spatiotemporal split-phase parallel feature extraction of CEUS video using R(2+1)D-18 convolutional neural network is used to construct a diagnostic model for HCC, and the information integrity and computational efficiency are balanced by a frame sampling strategy (15/30/45 frames), supplemented by gradient-weighted class activation mapping (Grad-CAM) visualisation to explain the decision basis. The diagnostic efficacy of the model is evaluated using accuracy, sensitivity, specificity, positive predictive value, negative predictive value and receiver operating characteristic (ROC) curve.

A total of 176 LR-M patients from the First Affiliated Hospital of Sun Yat-sen University were divided into training and validation sets in a 7:3 ratio, while 46 LR-M patients from the First Affiliated Hospital of Guangxi Medical University served as the test set. When using 45-frame videos as input, with sensitivities of 91.67% and 84.21%, specificities of 96.43% and 81.48%, and area under curves (AUCs) of 0.940 and 0.828 in the validation and test sets. The model had a higher detection rate of HCC compared with the alpha-fetoprotein (AFP) index (validation set: 50.00%, test set: 31.58%; both P value < 0.05), and the combined use of the DL model with AFP could optimize the overall diagnostic efficacy.

The DL model with CEUS spatiotemporal perception enables accurate diagnosis of HCC in LR-M category liver lesions, demonstrating significantly superior performance over AFP. Moreover, the combined diagnosis of the both can reduce unnecessary liver biopsy.

Exploring the challenges of artifact detection in endoscopic videos through the perspective of long-tailed learning.

Through analysis of artifact distribution characteristics in endoscopic videos from the public EAD2020 dataset, we develop a long-tailed endoscopic artifact detection algorithm via a decoupled framework. The proposed multi-expert system strategically partitions head and tail classes into two distinct subsets, with domain-specific experts processing each group separately. We innovatively design a hierarchical expert aggregation mechanism and further integrate visual prompts to enhance cross-group discrimination capabilities, ultimately achieving a significant improvement in artifact detection performance as measured by the comprehensive mean Average Precision (mAP) metric.

Benefiting from a more powerful backbone network and advanced classifier design, the proposed method achieves significant improvements in endoscopic video analysis: mAP for sequential video frame detection increases from 0.440 to 0.491, while mAP for individual video frame detection rises from 0.339 to 0.384. Notably, our approach comprehensively outperforms existing ensemble methods across all artifact categories.

The proposed multi-expert framework for long-tailed endoscopic video artifact detection achieves balanced performance across artifact categories through grouping and multi-expert methods. This architecture significantly enhances detection performance for tail classes while maintaining head class accuracy, effectively mitigating the inherent class imbalance problem. This study provides a novel direction for medical image detection tasks, particularly in data-imbalanced scenarios.

To develop a highly sensitive and selective electrochemical probe based on molecular imprinting polymers (MIPs) technology for the precise detection of 25-hydroxyvitamin D₃ [25-(OH)D₃], aiming to meet the diagnostic needs of vitamin D metabolism-related diseases.

The Co₃O₄@reduced graphene oxide (rGO)@Pt@Au composite substrate was constructed by layer-by-layer modification of screen-printed electrode (SPE) to enhance conductivity. Molecularly imprinted cavities were successfully formed through electropolymerization of polypyrrole and template molecule elution, imparting specific recognition capabilities. By optimizing the scan rate, deposition cycles, incubation time, and temperature, the balance between membrane structure, electron transfer efficiency, and specific binding ability was achieved. The sensor's response performance was evaluated using square wave voltammetry (SWV), and its sensitivity, selectivity, and reproducibility were analyzed.

Scanning electron microscopy (SEM) characterization revealed that after template elution, clear imprinted cavities were formed on the electrode surface, which could detect the target in the range of 100 nmol/L to 1 mmol/L, with a detection limit of 26.76 nmol/L. This detection strategy showed no specific response to interference at 10 times the concentration, and the reproducibility of the electrodes across three batches, with a relative standard deviation of 3.2%, indicated the reliability of the method. The spiked experiments also show its good practical application ability.

This study successfully established a new method with high sensitivity, strong anti-interference ability, and excellent reproducibility, providing a novel approach for the clinical detection of 25-(OH)D₃.

To investigate the mechanism and effect of porous hollow iron oxide nanoparticles (HION) nanosensitizers (HION/VER) loaded with VER-155008 prepared to enhance the radiofrequency ablation on hepatocellular carcinoma tumors.

The construction of porous hollow iron oxide nanosensitizers and drug loading were investigated by transmission electron microscopy and zeta potential experiments. The in vitro anti-tumor ability of the sensitizers was examined by cell viability and cell live/dead staining experiments. The in vitro anti-tumor mechanism of the sensitizers was investigated by measuring the intracellular levels of reactive oxygen species (ROS), glutathione (GSH) and heat shock protein 70 (HSP70).

Through material characterization analysis, porous hollow iron oxide nanosensitizers have been successfully constructed with a particle size of about 100 nm. HION/VER reduced tumor cell survival to 50% at 56.89 μg/mL and 45℃. Live/dead staining assay further confirmed that a large number of tumor cells died, indicating that HION/VER has good anti-tumor ability in vitro. Analysis by reactive oxygen fluorescence probe and glutathione level showed that reactive oxygen fluorescence was significantly enhanced and GSH levels decreased, indicating that HION/VER nano-sensitizer could promote iron death through the mechanism of disrupting redox balance. Analyzed by HSP70 immunofluorescence assay, the cellular HSP70 expression was significantly reduced after HION/VER treatment, indicating that the nano-sensitizer increased the sensitivity of tumor cells to mild radiofrequency ablation by inhibiting HSP70.

After HION/VER was endocytosed by the tumor cells, the drug uptake in the cells was increased, and the inhibition of HSP70 increased the sensitivity of the tumor cells to mild radiofrequency ablation by disrupting the redox balance to enhance the iron death, and the effect of radiofrequency ablation was achieved by effectively increasing the sensitivity of the tumor cells to mild radiofrequency ablation.

To conduct an in-depth analysis of the biomechanical differences between high-flexion and conventional knee prostheses, providing references for optimizing prosthetic design and enhancing clinical outcomes.

Finite element analysis (FEA) was employed to investigate the mechanical disparities between high-flexion and conventional femoral component prostheses. 3D models of the femoral condyle and tibial insert were constructed and comprehensively analyzed in ANSYS. The contact area, maximum contact stress, and peak von Mises stress were compared under identical rotation centers at flexion angles of 0°, 15°, 30°, 60°, 90°, 120°, 135°, and 150°.

The study revealed comparable mechanical performance between high-flexion and conventional prostheses at flexion angles below 90°. However, in high-flexion regimes beyond 120°, the high-flexion prosthesis demonstrated superior performance in simulations, exhibiting larger contact areas and lower contact stresses, demonstrating the efficacy of its design.

The methodology and findings of this study provide valuable insights for the design and manufacturing of high-flexion knee prostheses.

To investigate the value of multislice spiral computed tomography (MSCT) haemodynamic parameters in evaluating the efficacy of chemotherapy in patients with advanced gastric cancer.

120 patients with advanced gastric cancer admitted to our hospital between October 2021 and October 2023 who met the criteria for natriuresis were selected as study subjects. Fluorouracil-based combined with platinum-based chemotherapy was used, once every 3 weeks, for a total of 8 cycles. After patients completed 8 cycles of chemotherapy, they were divided into effective group (54 cases) and ineffective group (66 cases) according to the solid tumour efficacy evaluation criteria. Scanning was performed using MSCT to compare the changes in haemodynamic parameters before and after chemotherapy, including blood flow (BF), blood volume (BV), permeability surface (PS), perfusion blood flow (PBV), etc.

When efficacy was assessed after completion of the first and eighth cycles of chemotherapy, the overall effectiveness of the 120 patients with advanced gastric cancer was 32.50% (39/120) and 45.00% (54/120), respectively. After 8 cycles of chemotherapy, 120 patients with advanced gastric cancer showed a significant decreasing trend in the haemodynamic parameters BF, BV, PS and PBV by MSCT scanning, the difference was statistically significant (P<0.001). Before chemotherapy, the BF, PS and PBV of the effective group were higher than those of the ineffective group, and the BV was lower than those of the ineffective group. Although there were numerical differences between the two groups in the parameters of BF, BV, PS and PBV, none of them was statistically significant (P>0.05). Meanwhile, the decrease of these parameters in the effective group was more significant than that in the ineffective group (P<0.001). Compared with the ineffective group, patients in the effective group showed a more significant decrease in BF, BV, PS and PBV values (P<0.001).

MSCT haemodynamic parameters as important indicators of tumour angiogenesis and microcirculatory status, which can comprehensively assess the effect of chemotherapy in patients with advanced gastric cancer.

To investigate the short-term modulatory effects of cerebellar theta burst stimulation (TBS) on motor sequence learning ability and motor cortical excitability in healthy individuals.

Ten medical interns from the Department of Rehabilitation and Medicine at Ruijin Rehabilitation Hospital (March to June 2024) were enrolled in a single-group pretest-posttest study. Participants received a single session of continuous theta burst stimulation (cTBS) over the right cerebellum. Before and after stimulation, participants performed a serial reaction time task (SRTT) with their right hand, with reaction times (RTs) and accuracy rates recorded. Changes in corticospinal excitability were assessed by measuring motor evoked potential (MEP) amplitude, latency, resting motor threshold (RMT), intracortical inhibition (ICI), intracortical facilitation (ICF), and cortical silent period (CSP).

Post-cTBS, RTs for repeated sequences were prolonged (P<0.05), while RTs for random sequences shortened (P<0.05). Overall accuracy remained stable (P>0.05). In the left primary motor area (M1), MEP amplitude decreased (P<0.05), RMT increased (P<0.05), ICI inhibitory effect was enhanced (P<0.05), and CSP showed a trend toward prolongation (P>0.05). No significant changes were observed in ICF.

Cerebellar cTBS impairs the immediate performance of healthy participants during motor sequence learning, resulting in delayed responses, indicating that cerebellar cTBS may exert an adverse effect on immediate sequence consolidation. Concurrently, cerebellar cTBS reduces cortical excitability in the contralateral M1, consistent with the notion that the cerebellum exerts an inhibitory influence over cortical motor areas. These findings provide evidence for the specific role of cerebellar TBS in motor sequence learning.

To explore the effect of electroacupuncture combined with Shujin Zhuanggu decoction on bone mineral density and lumbar function of patients after percutaneous kyphoplasty (PKP).

A retrospective study was conducted, and 76 patients with vertebral compression fractures treated with PKP from January 2022 to November 2023 in our hospital were selected and divided into two groups according to different intervention methods. The control group (38 cases) was given Shujin Zhuanggu decoction after operation, and the observation group (38 cases) was given electroacupuncture stimulation on the basis of this. Bone mineral density, blood calcium, and serum β-C-terminal cross-linking telopeptide of type Ⅰ collagen (β-CTx) levels, lumbar function [Oswestry disability index (ODI) and Japanese Orthopaedic Association score (JOA)] and osteoporosis symptom grading were compared between the two groups.

Three months after the intervention, the bone mineral density and blood calcium levels of both groups were higher than those before the intervention, but the observation group was significantly higher than the control group. The serum β-CTx level decreased, and the observation group was lower than the control group, with statistical significance (P<0.05). Three months after the intervention, the ODI score of the two groups decreased and the JOA score increased, and the ODI score of the observation group was lower than that of the control group, and the JOA score was higher than that of the control group, with statistical significance (P<0.05). Three months after the intervention, the scores of symptoms in the two groups decreased compared with those before the intervention, and the scores of low back pain, weakness of waist and knees, weakness of lower limbs, pain of lower limbs and difficulty in walking in the observation group were lower than those in the control group, with statistical significance (P<0.05).

The effect of electroacupuncture combined with Shujin Zhuanggu decoction in patients after PKP is better than that of Shujin Zhuanggu decoction alone, which can improve patients' bone density, promote fracture healing and contribute to the recovery of lumbar function.

To investigate the application effect of the “triple-low” protocol (low tube voltage, low contrast agent dose, and low injection rate) in left atrial computed tomography angiography (CTA) and evaluate its feasibility and value in reducing radiation dose and contrast agent usage while ensuring image quality.

Sixty patients who underwent left atrial CTA examinations at the Radiology Department of the First Affiliated Hospital of Nanjing Medical University from June 2024 to December 2024 were prospectively enrolled and randomly divided into a conventional protocol group (n=30) and a “triple-low” protocol group (n=30). The conventional group used 120 kV tube voltage, 0.9 mL/kg contrast agent dose, 5.0 mL/s injection rate, and filtered back projection (FBP) algorithm. The “triple-low” group used 100 kV tube voltage, 0.7 mL/kg contrast agent dose, 4.0 mL/s injection rate, and hybrid iterative reconstruction (HIR) algorithm. Subjective image scores, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), effective radiation dose (ED), and iodine intake were compared between the two groups.

Both groups met diagnostic standards for subjective image quality, with no significant intergroup differences (P>0.05). The “triple-low” group showed significantly higher mean CT values in the left atrium compared to the conventional group (P<0.01), but there were no significant differences in SNR and CNR (P>0.05). The “triple-low” group exhibited significantly lower iodine intake and effective radiation dose than the conventional group (P<0.01).

The “triple-low” protocol significantly reduces radiation dose and contrast agent usage in left atrial CTA while maintaining image quality, offering a superior imaging solution for patients and demonstrating promising clinical value.

To investigate the impact of combining image-guided radiotherapy (IGRT) with a bladder volume measurement device on reducing positional errors in patients with locally advanced cervical cancer, and to provide references for precise clinical radiotherapy implementation.

Forty patients with locally advanced cervical cancer were randomly divided into a study group (using a bladder volume measurement device) and a control group (self-regulated bladder filling). Both groups underwent pre-treatment CT simulation (pCT) and daily cone-beam CT (CBCT) scans. Bladder contours were delineated on all pCT and CBCT images to obtain volumetric data. The accuracy of the bladder volume measurement device was assessed by comparing its readings with bladder volumes derived from pCT and CBCT. Inter-fraction bladder filling consistency and positional correction errors were analyzed by comparing CBCT-derived bladder volumes with pCT baselines. Irradiated volumes of the eighth region and the vital organs were compared between the two groups.

No significant differences were observed between bladder volumes measured by the device and those from pCT or CBCT (P = 0.539, P = 0.343, respectively). In the study group, no significant difference existed between pCT and CBCT bladder volumes (P=0.097), whereas the control group showed significant discrepancies (P<0.05). Positional correction errors in the X, Y, and Z axes were significantly smaller in the study group compared to the control group (P<0.05). Significant differences in irradiated volumes of the bladder and small bowel were observed between groups (P<0.05), but not for the CTV, PTV, rectum, or femoral heads (P>0.05).

The bladder volume measurement device effectively ensures consistent bladder filling during radiotherapy, reduces positional errors, stabilizes target dose coverage, and lowers radiation risks to organs at risk (e.g., small bowel). This approach enhances precision in hypofractionated radiotherapy for cervical cancer.

To analyze the influencing factors of restenosis after coronary stenting.

The clinical data of 194 patients with coronary atherosclerotic heart disease treated in our hospital from February 2020 to February 2023 were retrospectively analyzed. According to the coronary digital subtraction angiography (DSA), they were divided into restenosis group (23 cases) and non-restenosis group (171 cases) after coronary stenting. The clinical data and serological indexes of the two groups were compared, and the influencing factors were analyzed by multivariate Logistic regression, and the receiver operating characteristic (ROC) curve was drawn to analyze the predictive value.

Multivariate Logistic regression analysis showed that the number of diseased branches, diabetes history, number of stents ≥3, stent length≥30 mm, series stents, and high serum soluble triggering receptor expressed on myeloid cell-1 (sTREM-1) level were the independent influencing factors of restenosis after coronary stenting (OR=1.672, 1.624, 1.598, 1.384, 1.443, 1.579, P<0.05). The ROC curve showed that when logit (P)>12.746, the area under curve (AUC) value was 0.796, 95%CI was 0.752~0.835, χ² was 9.485, the diagnostic sensitivity was 80.00%, and the specificity was 66.67%.

The Logistic influencing factor model has high predictive value for restenosis after coronary stenting, so that the clinic can take corresponding intervention measures for patients with coronary atherosclerotic heart disease after coronary stenting as soon as possible to reduce the risk of restenosis after coronary stenting.

To study the predictive value of CT perfusion imaging (CTP) parameters for collateral circulation and functional outcomes after intravenous thrombolysis in patients with acute cerebral infarction (ACI).

A total of 100 patients with ACI admitted to the hospital from January 2021 to November 2024 were selected as the research subjects. All of them underwent CT perfusion imaging examination and received intravenous thrombolytic therapy after admission. The patients were divided into the good collateral group (46 cases) and the poor collateral group (54 cases) using digital subtraction angiography (DSA) as the gold standard. Compare the CTP parameters of the two groups of patients. According to the modified Rankin scale (mRS) score 90 days after the onset of the disease, the patients were divided into the good functional outcome group (65 cases) and the poor functional outcome group (35 cases). The predictive efficacy was analyzed using the receiver operating characteristic (ROC) curve.

Compared with the poor collateral group, the good collateral group had higher levels of cerebral blood flow (CBF) and cerebral blood volume (CBV) (P<0.05). The levels of mean transit time (MTT) and time to peak (TTP) were lower (P<0.05). The ROC curve results showed that CBF, CBV, MTT, TTP and the combined detection all had predictive efficacy for collateral circulation in patients (P<0.05), with AUC being 0.857, 0.800, 0.814, 0.713 and 0.931 respectively. Among the individual detections, CBF had the highest predictive efficacy but was still lower than the combined predictive efficacy. The Z values of CBF, CBV, MTT and TTP for the combined detection were calculated to be 1.37 (P>0.05), 2.21 (P<0.05), 1.71 (P>0.05) and 3.54 (P<0.05) respectively. The ROC curve results showed that CBF, CBV, MTT, TTP and the combined detection all had efficacy for the functional outcome of patients after intravenous thrombolysis (P<0.05), with AUC of 0.796, 0.811, 0.718, 0.784 and 0.902 respectively. Among the individual detects, CBV had the highest efficacy but was still lower than the combined predictive efficacy. The Z values of CBF, CBV, MTT and TTP for the combined detection were calculated to be 1.91 (P>0.05), 1.65 (P>0.05), 3.14 (P<0.05) and 2.13 (P<0.05) respectively.

In ACI patients, the levels of CBF and CBV in the good collateral circulation group were higher than those in the poor collateral circulation group, while the levels of MTT and TTP were lower. The CTP parameters in the good functional outcome group showed a similar trend to those in the good collateral circulation group. ROC analysis revealed that the CTP parameters had predictive efficacy for good collateral circulation and functional outcomes after intravenous thrombolysis.

To explore the application effect of microneedle radiofrequency acne treatment device in the treatment of acne, in order to guide the formulation of clinical treatment measures and improve the treatment effect on acne patients.

According to the random number table method, 513 acne patients admitted to the hospital from November 2021 to October 2022 were divided into an observation group (257 cases, treated with microneedle radiofrequency acne treatment device combined with minocycline hydrochloride) and a control group (256 cases, treated with minocycline hydrochloride), both treated for 12 weeks. Evaluate the clinical efficacy of two groups, compare the acne situation and quality of life before and after treatment, and statistically analyze the adverse reactions of the two groups.

The results showed that compared with the control group (67.97%), the observation group had a higher total effective rate (85.60%), with statistical differences (P<0.05). After treatment, both groups showed a decrease in the general acne grading system (GAGS) score, with the observation group having a lower score (P<0.05). The acne specific quality of life (Acne QoL) scores increased in both groups after treatment, with the observation group having a higher score (P<0.05). There was no statistically significant difference in adverse reactions between the two groups (P>0.05).

The microneedle radiofrequency acne treatment device has significant therapeutic effects on acne, can improve patients' quality of life, alleviate clinical symptoms, and has a certain degree of safety.

Proposes the construction of an intelligent scheduling system for medical technology, utilizing a data integration platform to optimize queuing order. The system aims to shorten patients' waiting times and enhance the efficiency of tests and examinations, ultimately improving the quality of medical services and the experience for patients.

This study utilizes Sybase PowerBuilder development tools in conjunction with MS SQL Server as the backend database. It adopts a Client/Server (C/S) architecture mode and is based on a data integration platform. Following the Web Service interface mode and adhering to XML standards, we aim to construct an intelligent scheduling system for medical technology, including infrastructure layer, data resource layer, platform application layer, data interaction layer, access control layer, application system layer and terminal access layer. This intelligent scheduling system is designed to fulfill business requirements while maintaining principles of stability, efficiency, convenience, and intelligence, and is seamlessly integrated with the hospital's internal data integration platform.

The system mainly consists of five functional: the front desk registration module, the examination room call module, the waiting area terminal module, the intelligent rule setting module, and the data query and statistics module. The practical application of the system has a significant impact, evidenced by improved registration times and user satisfaction compared to the period before the system's implementation. Statistically significant differences were observed (t=103.678, P<0.001; χ²=7.307, P<0.05), indicating that the system effectively reduces patient waiting times and improves physicians' work efficiency.

The medical intelligent scheduling system developed in this study enhances work efficiency and patient satisfaction by optimizing the processes of testing and examination. Furthermore, it offers a novel strategy for the development of smart hospitals.