Objective To explore the effectiveness of computer-aided technology in the treatment of primary elbow osteoarthritis combined with stiffness under arthroscopy. Methods The clinical data of 32 patients with primary elbow osteoarthritis combined with stiffness between June 2018 and December 2020 were retrospectively analyzed. There were 22 males and 10 females with an average age of 53.4 years (range, 31-71 years). X-ray film and three-dimensional CT examinations showed osteophytes of varying degrees in the elbow joint. Loose bodies existed in 16 cases, and there were 7 cases combined with ulnar nerve entrapment syndrome. The median symptom duration was 2.5 years (range, 3 months to 22.5 years). The location of bone impingement from 0° extension to 140° flexion of the elbow joint was simulated by computer-aided technology before operation and a three-dimensional printed model was used to visualize the amount and scope of impinging osteophytes removal from the anterior and posterior elbow joint to accurately guide the operation. Meanwhile, the effect of elbow joint release and impinging osteophytes removal was examined visually under arthroscopy. The visual analogue scale (VAS) score, Mayo elbow performance score (MEPS), and elbow range of motion (extension, flexion, extension and flexion) were compared between before and after operation to evaluate elbow function. Results The mean operation time was 108 minutes (range, 50-160 minutes). All 32 patients were followed up 9-18 months with an average of 12.5 months. There was no other complication such as infection, nervous system injury, joint cavity effusion, and heterotopic ossification, except 2 cases with postoperative joint contracture at 3 weeks after operation due to the failure to persist in regular functional exercises. Loose bodies of elbow and impinging osteophytes were removed completely for all patients, and functional recovery was satisfactory. At last follow-up, VAS score, MEPS score, extension, flexion, flexion and extension range of motion significantly improved when compared with preoperative ones (P<0.05). Conclusion Arthroscopic treatment of primary elbow osteoarthritis combined with stiffness using computer-aided technology can significantly reduce pain, achieve satisfactory functional recovery and reliable effectiveness.
ObjectiveTo explore the clinical value of computer-assisted surgical planning in the treatment of ankle fractures. MethodsBetween January 2012 and January 2014, open reduction and internal fixation were performed on 42 patients with ankle fractures. There were 22 males and 20 females with an average age of 52 years (range, 19-72 years). The causes were spraining injury (20 cases), traffic accident injury (14 cases), and falling from height injury (8 cases). The time from injury to operation was 5 hours to 12 days (mean, 2.5 days). All fractures were closed trimalleolar fractures. According to Lauge-Hansen classification, 25 cases were rated as supination extorsion type IV, 13 as pronation extorsion type IV, and 4 as pronation abduction type Ⅲ. The preoperative planning was made by virtual reduction and internal fixation using Superimage software. ResultsThe mean operation time was 93.7 minutes (range, 76-120 minutes). Delayed wound healing occurred in 1 case, and secondary healing was obtained after treatment; primary healing of incision was achieved in the other patients. Postoperative X-ray films and CT images showed anatomic reduction of fracture and good position of internal fixation. All patients were followed up 14.6 months on average (range, 9-27 months). The range of motion of the affected ankle was close to the normal side at 6-8 weeks. The mean fracture healing time was 13.1 weeks (range, 11-17 weeks). Degenerative change of the ankle joint was observed in 3 cases (7.1%) with manifestation of mild narrowing of joint space on the X-ray films at last follow-up. According to Baird-Jackson score system, the results were excellent in 24 cases, good in 13 cases, and fair in 5 cases, with an excellent and good rate of 88%. ConclusionComputer-assisted surgical planning for ankle fractures can help surgeons identify type of ankle fractures and improve surgical scheme for guiding fracture reduction and selecting and placing implants, so good effectiveness can be obtained.
Early diagnosis and treatment of colorectal polyps are crucial for preventing colorectal cancer. This paper proposes a lightweight convolutional neural network for the automatic detection and auxiliary diagnosis of colorectal polyps. Initially, a 53-layer convolutional backbone network is used, incorporating a spatial pyramid pooling module to achieve feature extraction with different receptive field sizes. Subsequently, a feature pyramid network is employed to perform cross-scale fusion of feature maps from the backbone network. A spatial attention module is utilized to enhance the perception of polyp image boundaries and details. Further, a positional pattern attention module is used to automatically mine and integrate key features across different levels of feature maps, achieving rapid, efficient, and accurate automatic detection of colorectal polyps. The proposed model is evaluated on a clinical dataset, achieving an accuracy of 0.9982, recall of 0.9988, F1 score of 0.9984, and mean average precision (mAP) of 0.9953 at an intersection over union (IOU) threshold of 0.5, with a frame rate of 74 frames per second and a parameter count of 9.08 M. Compared to existing mainstream methods, the proposed method is lightweight, has low operating configuration requirements, high detection speed, and high accuracy, making it a feasible technical method and important tool for the early detection and diagnosis of colorectal cancer.
Objective To investigate the value of computer-aided design (CAD) in defining the resection boundary, reconstructing the pelvis and hip in patients with pelvis tumors. Methods Between November 2006 and April 2009, 5 cases of pelvis tumors were treated surgically using CAD technology. There were 3 males and 2 females with an average age of 36.4 years (range, 24-62 years). The cause was osteosarcoma, giant cell tumor of bone, and angiosarcoma in 1 case, respectively,and chondrosarcoma in 2 cases. According to the Enneking system for staging benign and mal ignant musculoskeletal tumors, regions I, I + II, III, IV, and I + IV is in 1 case, respectively. According to the principle of reverse engineering, 5 patients with pelvis tumors were checked with lamellar CT/MRI scanning, whose two-dimensional data were obtained in disease area. The three-dimensional reconstruction of pelvic anatomical model, precise resection boundary of tumor, individual surgical template, individual prosthesis, and surgical simulation were precisely made by computer with CAD software. Based on the proposal of CAD, the bone tumor was resected accurately, and allograft il ium with internal fixation instrument or allogeneic il ium with personal ized prosthetic replacement were used to reconstruct the bone defect after tumor was resected. Results The operation was successfully performed in 5 cases. The average operation time was 7.9 hours, and the average blood loss was 3 125 mL. Hemorrhage and cerebrospinal fluid leakage occurred in 1 case, respectively, and were cured after debridement. Five patients were followed up from 24 to 50 months (mean, 34.5 months). All patients began non-weight bearing walk with double crutches at 4-6 weeks after operation, and began walk at 3-6 months after operation. Local recurrence developed in 2 patients at 18 months after operation, and resection and radiotherapy were performed. According to International Society of Limb Salvage criteria for curative effectiveness of bone tumor l imb salvage, the results were excellent in 2 and good in 3. Conclusion The individual surgical template, individual prosthesis, and surgical simulation by CAD ensure the precision and rel iabil ity of pelvis tumors resection. The CAD technology promotes pelvis tumor resection and the reconstruction of pelvis to individual treatment stage, and good curative effectiveness can be obtained.
Objective To introduce the recent advances of the application of computer technology in tissue engineering. Methods The recent original articlesrelated to computer technology, medical image technology, computer-aided design, the advanced manufacture technology were summarized and systematically analyzed.Results Computer-aided tissue engineering is a new fieldon tissue engineering. It is the future direction of tissue engineering study. This article reviews recent development of medical CT/MRI scanning, three-dimensional reconstruction, anatomical modeling, computeraided design, computer-aided manufacturing, rapid prototyping， RP manufacturing of tissue engineering scaffolds and computeraided implantation.Conclusion Computer-aided tissue engineering can be used in scaffolds design and fabrication, computer-aided artificial tissue implantation. It is a new field on tissue engineering.
Medical visual question answering (MVQA) plays a crucial role in the fields of computer-aided diagnosis and telemedicine. Due to the limited size and uneven annotation quality of the MVQA datasets, most existing methods rely on additional datasets for pre-training and use discriminant formulas to predict answers from a predefined set of labels. This approach makes the model prone to overfitting in low resource domains. To cope with the above problems, we propose an image-aware generative MVQA method based on image caption prompts. Firstly, we combine a dual visual feature extractor with a progressive bilinear attention interaction module to extract multi-level image features. Secondly, we propose an image caption prompt method to guide the model to better understand the image information. Finally, the image-aware generative model is used to generate answers. Experimental results show that our proposed method outperforms existing models on the MVQA task, realizing efficient visual feature extraction, as well as flexible and accurate answer outputs with small computational costs in low-resource domains. It is of great significance for achieving personalized precision medicine, reducing medical burden, and improving medical diagnosis efficiency.
Early screening is an important means to reduce breast cancer mortality. In order to solve the problem of low breast cancer screening rates caused by limited medical resources in remote and impoverished areas, this paper designs a breast cancer screening system aided with portable ultrasound Clarius. The system automatically segments the tumor area of the B-ultrasound image on the mobile terminal and uses the ultrasound radio frequency data on the cloud server to automatically classify the benign and malignant tumors. Experimental results in this study show that the accuracy of breast tumor segmentation reaches 98%, and the accuracy of benign and malignant classification reaches 82%, and the system is accurate and reliable. The system is easy to set up and operate, which is convenient for patients in remote and poor areas to carry out early breast cancer screening. It is beneficial to objectively diagnose disease, and it is the first time for the domestic breast cancer auxiliary screening system on the mobile terminal.
Objective To compare the effect of three-dimensional visual (3DV) model, three-dimensional printing (3DP) model and computer-aided design (CAD) modified 3DP model in video-assisted thoracoscopic surgery (VATS) sublobular resection. MethodsThe clinical data of patients who underwent VATS sublobular resection in the Affiliated Hospital of Hebei University from November 2021 to August 2022 were retrospectively analyzed. The patients were divided into 3 groups including a 3DV group, a 3DP group and a CAD-3DP group according to the tools used. The perioperative indexes and subjective evaluation of operators, patients and their families were compared. ResultsA total of 22 patients were included. There were 5 males and 17 females aged 32-77 (56.95±12.50) years. There were 9 patients in the 3DV group, 6 patients in the 3DP group, and 7 patients in the CAD-3DP group. There was no statistical difference in the operation time, intraoperative blood loss, drainage volume, hospital stay time or postoperative complications among the groups (P>0.05). Based on the subjective evaluations of 4 surgeons, the CAD-3DP group was better than the 3DV group in the preoperative planning efficiency (P=0.025), intuitiveness (P=0.045) and doctor-patient communication difficulty (P=0.034); the CAD-3DP group was also better than the 3DP group in the overall satisfaction (P=0.023), preoperative planning difficulty (P=0.046) and efficiency (P=0.014). Based on the subjective evaluations of patients and their families, the CAD-3DP group was better than the 3DP group in helping understand the vessel around the tumor (P=0.016), surgical procedure (P=0.020), procedure selection (P=0.029), and overall satisfaction (P=0.048); the CAD-3DP group was better than the 3DV group in helping understand the tumor size (P=0.038). ConclusionCAD-modified 3DP model has certain advantages in pre-planning, intraoperative navigation and doctor-patient communication in the VATS sublobectomy.
Objective To improve the fitness and initial fixation strength between the hip and bone and to optimize the shape of the prosthetic implants. Methods The cross-section of hip canal was automatically extracted by Image processing. By using taper curve fit,hypocurve predigesting and the curve of shape center fit, the parameters of long-short diameter and the curve of shape center were got to design the hip shape.CAD was adopted to analyze and evaluate the configuration of bone and shape of hip.The “peg-in-hole” was employed to optimize the shape of implant by the visual test of “Drawingout” in computer. Results 23.2% hip-bone average matching rate and 0.033% bone damage rate were presented by CAD analysis. The implant extraction path were validated visually and quantitatively by measuring the maximum amount of overlap in the path configuration. Conclusion The valuable method for prothsetic hip design was presented by the way of image processing,graphics design and optimizingdesign in this study.
ObjectivesTo develop a fundus photography (FP) image lesion recognition model based on the EfficientNet lightweight convolutional neural network architecture, and to preliminary evaluate its recognition performance. MethodsA diagnostic test. The data was collected in the Department of Ophthalmology at Sichuan Provincial People's Hospital from June 2023 to June 2025. A lightweight 16-category lesion recognition model was constructed based on deep learning and 610 072 FP images. The FP images were sourced from Sichuan Provincial People's Hospital as well as the APTOS, Diabetic Retinopathy_2015, Diabetic Retinopathy_2019, and Retinal Disease datasets. Model performance was evaluated as follows: first, testing was performed on four independent external validation sets using metrics such as accuracy, F1 score (the harmonic mean of precision and recall), and the area under the receiver operating characteristic curve (AUC) to measure the model's generalizability and accuracy. Second, the classification results of the model were compared with those of junior and mid-level ophthalmologists (two each) using the overlapping confidence interval (CI) comparison method to assess the clinical experience level corresponding to the model's medical proficiency. ResultsThe model achieved an accuracy of 96.78% (59 039/61 003), an F1 score of 82.51% (50 334/61 003), and an AUC of 99.93% (60 960/61 003) on the validation set. On the four external validation sets, it achieved an average accuracy of 87.77% (57 358/65 350), an average precision of 87.06% (56 894/65 350), and an average Kappa value of 82.28%. The average accuracy of FP image lesion identification for junior and mid-level ophthalmologists was 79.00% (79/100) (95%CI 67.71-90.29) and 87.00% (87/100) (95%CI 77.68-96.32), respectively. ConclusionsA 16-category FP image lesion recognition model is successfully constructed based on the EfficientNet lightweight convolutional neural network architecture. Its clinical performance preliminarily reaches the level of mid-level ophthalmologists.