To address the current issues of data imbalance and scarcity in photoplethysmography (PPG) data for type 2 diabetes mellitus (T2DM) prediction, this study proposes an improved conditional Wasserstein generative adversarial network with gradient penalty (CWGAN-GP). The algorithm integrated gated recurrent unit (GRU) networks and self-attention mechanisms to construct a generator, aiming to produce high-quality PPG signals. Various data augmentation methods, including the improved CWGAN-GP, were employed to expand the PPG dataset, and multiple classifiers were applied for T2DM prediction analysis. Experimental results showed that the model trained on data generated by the improved CWGAN-GP achieved the optimal prediction performance. The highest accuracy reached 0.895 0, and compared with other data enhancement methods, this approach exhibited significant advantages in terms of precision and F1-score. The generated data notably enhances the accuracy and generalization ability of T2DM prediction models, providing a more reliable technical basis for non-invasive early T2DM screening based on PPG signals.
Photoplethysmography/methods* ; Diabetes Mellitus, Type 2/diagnosis* ; Humans ; Algorithms ; Neural Networks, Computer ; Signal Processing, Computer-Assisted ; Prediction Algorithms

Heart sounds are critical for early detection of cardiovascular diseases, yet existing studies mostly focus on traditional signal segmentation, feature extraction, and shallow classifiers, which often fail to sufficiently capture the dynamic and nonlinear characteristics of heart sounds, limit recognition of complex heart sound patterns, and are sensitive to data imbalance, resulting in poor classification performance. To address these limitations, this study proposes a novel heart sound classification method that integrates improved Mel-frequency cepstral coefficients (MFCC) for feature extraction with a convolutional neural network (CNN) and a deep Transformer model. In the preprocessing stage, a Butterworth filter is applied for denoising, and continuous heart sound signals are directly processed without segmenting the cardiac cycles, allowing the improved MFCC features to better capture dynamic characteristics. These features are then fed into a CNN for feature learning, followed by global average pooling (GAP) to reduce model complexity and mitigate overfitting. Lastly, a deep Transformer module is employed to further extract and fuse features, completing the heart sound classification. To handle data imbalance, the model uses focal loss as the objective function. Experiments on two public datasets demonstrate that the proposed method performs effectively in both binary and multi-class classification tasks. This approach enables efficient classification of continuous heart sound signals, provides a reference methodology for future heart sound research for disease classification, and supports the development of wearable devices and home monitoring systems.
Heart Sounds/physiology* ; Humans ; Algorithms ; Neural Networks, Computer ; Signal Processing, Computer-Assisted ; Deep Learning ; Cardiovascular Diseases/diagnosis* ; Classification Algorithms

Lung cancer is a leading cause of cancer-related deaths worldwide, with its high mortality rate primarily attributed to delayed diagnosis. Radiomics, by extracting abundant quantitative features from medical images, offers novel possibilities for early diagnosis and precise treatment of lung cancer. This article reviewed the latest advancements in radiomics for lung cancer management, particularly its integration with artificial intelligence (AI) to optimize diagnostic processes and personalize treatment strategies. Despite existing challenges, such as non-standardized image acquisition parameters and limitations in model reproducibility, the incorporation of AI significantly enhanced the precision and efficiency of image analysis, thereby improving the prediction of disease progression and the formulation of treatment plans. We emphasized the critical importance of standardizing image acquisition parameters and discussed the role of AI in advancing the clinical application of radiomics, alongside future research directions.
Humans ; Lung Neoplasms/diagnosis* ; Artificial Intelligence ; Early Detection of Cancer/methods* ; Precision Medicine ; Image Processing, Computer-Assisted/methods* ; Tomography, X-Ray Computed ; Radiomics

Thyroid nodules are a common endocrine disorder, and their early detection and accurate diagnosis are crucial for the prevention of thyroid cancer. However, the highly heterogeneous morphology and boundaries of thyroid nodules pose significant challenges to their precise identification and classification. Traditional diagnostic approaches rely heavily on physicians' experience, which increases the risk of misdiagnosis and missed diagnoses. With the rapid advancement of computer-aided diagnosis (CAD) technologies, applying deep learning algorithms to the analysis of thyroid nodule ultrasound images has shown great potential. This paper reviews the latest research progress on deep learning-based CAD methods for thyroid nodules, with a focus on their applications in image preprocessing, segmentation and classification. The advantages and limitations of current techniques are analyzed, and potential future directions are discussed. This review aims to highlight the potential of deep learning in thyroid nodule diagnosis and to provide a foundation for selecting feasible pathways for future clinical applications.
Humans ; Thyroid Nodule/diagnostic imaging* ; Deep Learning ; Ultrasonography/methods* ; Diagnosis, Computer-Assisted/methods* ; Algorithms ; Thyroid Neoplasms/diagnostic imaging* ; Image Processing, Computer-Assisted/methods*

Dynamic navigation technology can "real-time guide" the implantologist to place the implant in the alveolar bone of the missing tooth area according to the preoperative design of the optimal site and path, making the whole implant surgery process more safe and precise. In order to further promote the standardized application of oral implant dynamic navigation technology, China Association of Gerontology and Geriatrics has convened distinguished experts to engage in deliberations and develop the standard. This standard covers the basic requirements, indications and contraindications, operation procedures, common complications and treatment measures, and accuracy verification. This standard can be used as a reference for the use of dynamic navigation technology in implant surgery.
Humans ; Dental Implantation, Endosseous/standards* ; Surgery, Computer-Assisted/standards* ; Dental Implants ; Surgical Navigation Systems/standards*

OBJECTIVE: To evaluate the early effectiveness of navigation-free robot-assisted total knee arthroplasty (TKA) compared to traditional TKA in the treatment of knee osteoarthritis combined with extra-articular deformities. METHODS: The clinical data of 30 patients with knee osteoarthritis combined with extra-articular deformities who met the selection criteria between June 2019 and January 2024 were retrospectively analyzed. Fifteen patients underwent CORI navigation-free robot-assisted TKA and intra-articular osteotomy (robot group) and 15 patients underwent traditional TKA and intra-articular osteotomy (traditional group). There was no significant difference in age, gender, body mass index, affected knee side, extra-articular deformity angle, deformity position, deformity type, and preoperative knee range of motion, American Knee Society (KSS) knee score and KSS function score, and lower limb alignment deviation between the two groups ( P>0.05). The operation time, intraoperative blood loss, and complications of the two groups were recorded and compared. The knee range of motion and lower limb alignment deviation were recorded before operation and at 6 months after operation, and the knee joint function was evaluated by KSS knee score and function score. RESULTS: There was no significant difference in operation time between the two groups ( P>0.05); the intraoperative blood loss in the robot group was significantly less than that in the traditional group ( P<0.05). Patients in both groups were followed up 6-12 months, with an average of 8.7 months. The incisions of all patients healed well, and there was no postoperative complication such as thrombosis or infection. At 6 months after operation, X-ray examination showed that the position of the prosthesis was good in both groups, and there was no loosening or dislocation of the prosthesis. The knee joint range of motion, the lower limb alignment deviation, and the KSS knee score and KSS function score significantly improved in both groups ( P<0.05) compared to preoperative ones. The changes of lower limb alignment deviation and KSS function score between pre- and post-operation in the robot group were significantly better than those in the traditional group ( P<0.05), while the changes of other indicators between pre- and post-operation in the two groups were not significant ( P>0.05). CONCLUSION Compared to traditional TKA, navigation-free robot-assisted TKA for knee osteoarthritis with extra-articular deformities results in less intraoperative blood loss, more precise reconstruction of lower limb alignment, and better early effectiveness. However, long-term effectiveness require further investigation.
Humans ; Arthroplasty, Replacement, Knee/methods* ; Osteoarthritis, Knee/surgery* ; Robotic Surgical Procedures/methods* ; Male ; Female ; Retrospective Studies ; Range of Motion, Articular ; Middle Aged ; Aged ; Treatment Outcome ; Osteotomy/methods* ; Knee Joint/physiopathology* ; Operative Time

OBJECTIVE: To explore effectiveness of TiRobot-assisted screw implantation in the treatment of coracoid process fractures of the scapula. METHODS: A retrospective analysis was conducted on the clinical data from 24 patients with coracoid process fractures of the scapula admitted between September 2019 and January 2024 and met selection criteria. Among them, 12 patients underwent TiRobot-assisted screw implantation (robot group) and 12 underwent manual screw implantation (control group) during internal fixation. There was no significant difference ( P>0.05) in baseline data such as gender, age, body mass index, disease duration, cause of injury, coracoid process fracture classification, and proportion of patients with associated injuries between the two groups. The incision length, operation time, intraoperative blood loss, hospital stay, accuracy of screw placement, coracoid process fracture healing time, and complications were recorded and compared, as well as pain visual analogue scale (VAS) score, and Constant-Murley score at last follow-up. RESULTS: The intraoperative blood loss and incision length in the robot group were significantly lower than those in the control group ( P<0.05); however, there was no significant difference in operation time and hospital stay between the two groups ( P>0.05). All patients were followed up 8-27 months (mean, 17.5 months), and the difference in follow-up time between the two groups was not significant ( P>0.05). At last follow-up, the VAS score for shoulder pain in the robot group was signifncatly lower compared to the control group, and the Constant-Murley score was significantly higher ( P<0.05). In the robot group, 16 screws were implanted intraoperatively, while 13 screws were implanted in the control group. Radiographic re-evaluation showed that the excellent and good rate of screw implantation was higher in the robot group (93.8%, 15/16) than in the control group (61.5%, 8/13), but the difference in the precision of screw implantation between the two groups was not significant ( P>0.05). Four patients in the robot group and 1 in the control group achieved double screws fixation; however, the difference in achieving double screws fixation between the two groups was not significant ( P>0.05). All fractures healed in both groups with 1 case of malunion in the control group. There was no significant difference in healing time between the two groups ( P>0.05). During follow-up, 1 patient in the control group experienced screw loosening and displacement. There was no significant difference in the incidence of screw loosening and fracture malunion between the two groups ( P>0.05). CONCLUSION Compared with manual screw implantation, TiRobot-assisted minimally invasive treatment of coracoid process fractures of the scapula can reduce intraoperative blood loss, shorten incision length, alleviate pain, and obtain better promote shoulder joint functional recovery.
Humans ; Male ; Female ; Retrospective Studies ; Fracture Fixation, Internal/instrumentation* ; Minimally Invasive Surgical Procedures/instrumentation* ; Adult ; Middle Aged ; Fractures, Bone/surgery* ; Bone Screws ; Coracoid Process/surgery* ; Robotic Surgical Procedures/methods* ; Scapula/surgery* ; Treatment Outcome ; Operative Time ; Young Adult ; Length of Stay ; Blood Loss, Surgical

OBJECTIVE: To explore the advantages and effectiveness of the independently developed intelligent orthopedic robot-assisted distal locking of femoral intramedullary nails. METHODS: Thirty-two adult cadaveric femur specimens were randomly divided into two groups, with 16 specimens in each group. The experimental group used the intelligent orthopedic robot to assist in the distal locking of femoral intramedullary nail holes, while the control group used the traditional method of manual locking under X-ray fluoroscopy. The locking time, fluoroscopy times, and the success rate of first locking were recorded and compared between the two groups. RESULTS: The locking time of the experimental group was (273.94±38.67) seconds, which was shorter than that of the control group [(378.38±152.72) seconds], and number of fluoroscopies was (4.56±0.81) times, which was less than that of the control group [(8.00±3.98) times]. The differences were significant [ MD=73.054 (-37.187, 85.813), P=0.049; MD=1.969 (-1.437, 2.563), P=0.002]. The first locking success rate of the experimental group was 100% (16/16), which was significantly higher than that of the control group (68.75%, 11/16) ( P=0.043). CONCLUSION The efficiency of distal locking of femoral intramedullary nails assisted by the intelligent orthopedic robot is significantly higher than that of the traditional manual locking method under fluoroscopy, as it can markedly reduce the time required for distal locking of femoral intramedullary nails, decrease intraoperative radiation exposure, and increase the success rate of locking.
Humans ; Fracture Fixation, Intramedullary/instrumentation* ; Bone Nails ; Fluoroscopy ; Femur/diagnostic imaging* ; Femoral Fractures/surgery* ; Robotic Surgical Procedures/instrumentation* ; Cadaver ; Adult ; Robotics ; Male

OBJECTIVE: To investigate the effectiveness of digital three-dimensional (3D) printing osteotomy guide plate assisted total knee arthroplasty (TKA) in treatment of knee osteoarthritis (KOA) patients with femoral internal implants. METHODS: The clinical data of 55 KOA patients who met the selection criteria between July 2021 and October 2023 were retrospectively analyzed. Among them, 26 cases combined with femoral implants were treated with digital 3D printing osteotomy guide plate assisted TKA (guide plate group), and 29 cases were treated with conventional TKA (control group). There was no significant difference in gender, age, body mass index, side, Kellgren-Lawrence classification, preoperative visual analogue scale (VAS) score, Hospital for Special Surgery (HSS) knee score, knee range of motion, and other baseline data between the two groups ( P>0.05). The operation time, intraoperative blood loss, incision length, postoperative first ambulation time, surgical complications; VAS score, knee HSS score, knee range of motion before operation, at 1 week and 3 months after operation, and at last follow-up; distal femoral lateral angle, proximal tibial medial angle, hip-knee-ankle angle and other imaging indicators at last follow-up were recorded and compared between the two groups. RESULTS: The operation time, incision length, intraoperative blood loss, and postoperative first ambulation time in the guide plate group were significantly lower than those in the control group ( P<0.05). In the control group, there were 1 case of incision rupture and bleeding and 1 case of lower limb intermuscular venous thrombosis, which was cured after symptomatic treatment. There was no complication such as neurovascular injury, incision infection, or knee prosthesis loosening in both groups. Patients in both groups were followed up 12-26 months, with an average of 16.25 months. The VAS score, HSS score, and knee range of motion improved at each time point after operation in both groups, and further improved with time after operation, the differences were significant ( P<0.05). The above indicators in the guide plate group were significantly better than those in the control group at 1 week and 3 months after operation ( P<0.05), and there was no significant difference between the two groups at last follow-up ( P>0.05). At last follow-up, the distal femoral lateral angle, the proximal tibial medial angle, and the hip-knee-ankle angle in the guide plate group were significantly better than those in the control group ( P<0.05). CONCLUSION The application of digital 3D printing osteotomy guide plate assisted TKA in the treatment of KOA patients with femoral implants can simplify the surgical procedures, overcome limitations of conventional osteotomy guides, reduce surgical trauma, achieve individualized and precise osteotomy, and effectively restore lower limb alignment and knee joint function.
Humans ; Arthroplasty, Replacement, Knee/instrumentation* ; Osteoarthritis, Knee/surgery* ; Osteotomy/instrumentation* ; Male ; Retrospective Studies ; Female ; Printing, Three-Dimensional ; Femur/surgery* ; Middle Aged ; Bone Plates ; Range of Motion, Articular ; Aged ; Treatment Outcome ; Surgery, Computer-Assisted/methods* ; Knee Prosthesis ; Knee Joint/surgery* ; Operative Time

OBJECTIVE: To investigate the effectiveness of 5G remote robotic surgery in the treatment of pelvic fractures. METHODS: A retrospective analysis was conducted on the clinical data of 160 patients with pelvic fractures admitted between July 2023 and June 2024 who met the selection criteria. Among these patients, 80 underwent internal fixation surgery with the assistance of 5G remote robotic surgery (5G group), while 80 received local robotic surgical assistance (control group). Baseline characteristics, including gender, age, body mass index, disease duration, cause of injury, and fracture classification, were compared between the two groups, and no significant difference was found ( P>0.05). The incision length, operation time, intraoperative blood loss, hospital stay, accuracy of screw placement, maximum residual displacement postoperatively, quality of fracture reduction, incidence of complications, Majeed pelvic function score and classification at last follow-up were recorded and compared between the two groups. RESULTS: In the 5G group, 180 screws were implanted during surgery, while 213 screws were implanted in the control group. The 5G group demonstrated significantly reduced intraoperative blood loss and shorter incision length compared to the control group ( P<0.05). No significant difference was observed between the two groups in terms of operation time or hospital stay ( P>0.05). Radiographic evaluation revealed excellent and good reduction rates of 98.8% (79/80) in the 5G group and 97.5% (78/80) in the control group, while excellent and good screw placement accuracy rates were 98.3% (177/180) in the 5G group and 95.8% (204/213) in the control group. No significant difference was found between the two groups in maximum residual displacement, reduction quality, or screw placement accuracy ( P>0.05). All patients were followed up 7-16 months (mean, 11.3 months), with no significant difference in follow-up duration between the groups ( P>0.05). No perioperative or follow-up complication, such as wound infection, iatrogenic fractures, iatrogenic neurovascular injury, screw loosening or breakage, or nonunion, were observed in either group. The control group exhibited a worse degree of gait alteration compared to the 5G group ( P<0.05), while no significant difference was found in incidences of squatting limitation or persistent pain ( P>0.05). At last follow-up, no significant difference was observed between the groups in Majeed pelvic function scores or grading ( P>0.05). CONCLUSION Compared with the local surgery group, 5G remote robotic surgery supported by remote expert technical guidance demonstrated smaller incision lengths, less intraoperative blood loss, and fewer postoperative complications, and was shown to be a precise, minimally invasive, safe, and reliable surgical method.
Humans ; Robotic Surgical Procedures/instrumentation* ; Pelvic Bones/surgery* ; Male ; Retrospective Studies ; Fracture Fixation, Internal/instrumentation* ; Female ; Fractures, Bone/surgery* ; Middle Aged ; Bone Screws ; Adult ; Operative Time ; Treatment Outcome ; Blood Loss, Surgical ; Length of Stay ; Postoperative Complications/epidemiology* ; Aged ; Young Adult