Abstract： Diseases prognosis is often influenced by multiple factors, and some intricate non-linear relationships exist among those factors. Artificial neural network （ANN）, an artificial intelligence model, simulates the work mode of biological neurons and has a b capability to analyze multi-factor non-linear relationships. In recent years, ANN is increasingly applied in clinical medical fields, especially for the prediction of disease prognosis. This article focuses on the basic principles of ANN and its application in disease prognosis research.
Objective To evaluate the clinical valueof the revision of total hip replacement(THR), to analyse the reason of the rev isions, and to discuss the main difficulties and measures to manage it.Methods From June 1998 to January 2002, 15 cases (15 hips) were revised on totalhip replacement. The reasons for revision in the cases were as follows:failure of primary operative techenique, loosening and sinking of the components, displacement of the prosthesis, erosion of the acetabulum, as well as fracture of the femoral stem. The main difficulties of the revision were:poor health condition of the patients; the remove of the prosthesis of the primary THR,especially the broken femoral stem and the cements; the loss of localbone. The measures to remove the broken femoral stem were described.ResultsAll cases were followed up 2.4 years on average: 2 patients died from heart disease and cerebrovascular disease respectively, while the good results were achieved in the others.No infection, dislocation, loosening, and other complications occurred. The good effects were related with following factors:mild degree of illness; no severe bone defect; most of the first femoral head replacement.Conclusion The revision of THRis a more difficult operation, so that the special instrument and equipment andoperative experience are required.
ObjectiveTo develop an artificial intelligence based three-dimensional (3D) preoperative planning system (AIHIP) for total hip arthroplasty (THA) and verify its accuracy by preliminary clinical application.MethodsThe CT image database consisting of manually segmented CT image series was built up to train the independently developed deep learning neural network. The deep learning neural network and preoperative planning module were assembled within a visual interactive interface—AIHIP. After that, 60 patients (60 hips) with unilateral primary THA between March 2017 and May 2020 were enrolled and divided into two groups. The AIHIP system was applied in the trial group (n=30) and the traditional acetate templating was applied in the control group (n=30). There was no significant difference in age, gender, operative side, and Association Research Circulation Osseous (ARCO) grading between the two groups (P>0.05). The coincidence rate, preoperative and postoperative leg length discrepancy, the difference of bilateral femoral offsets, the difference of bilateral combined offsets of two groups were compared to evaluate the accuracy and efficiency of the AIHIP system.ResultsThe preoperative plan by the AIHIP system was completely realized in 27 patients (90.0%) of the trial group and the acetate templating was completely realized in 17 patients (56.7%) of the control group for the cup, showing significant difference (P<0.05). The preoperative plan by the AIHIP system was completely realized in 25 patients (83.3%) of the trial group and the acetate templating was completely realized in 16 patients (53.3%) of the control group for the stem, showing significant difference (P<0.05). There was no significant difference in the difference of bilateral femoral offsets, the difference of bilateral combined offsets, and the leg length discrepancy between the two groups before operation (P>0.05). The difference of bilateral combined offsets at immediate after operation was significantly less in the trial group than in the control group (t=?2.070, P=0.044); but there was no significant difference in the difference of bilateral femoral offsets and the leg length discrepancy between the two groups (P>0.05).ConclusionCompared with the traditional 2D preoperative plan, the 3D preoperative plan by the AIHIP system is more accurate and detailed, especially in demonstrating the actual anatomical structures. In this study, the working flow of this artificial intelligent preoperative system was illustrated for the first time and preliminarily applied in THA. However, its potential clinical value needs to be discovered by advanced research.
The primary results of five patients in whomthe block hydroxyapatite artificial bone (BHAB)used in maxillofacial plastic repair were reported. All incisions healed up with no evidence ofinfection. None of the implants was rejected norhad resorption changes. Satisfactory estheticaleffects were maintained. The results demonst-rated BHAB had a good biocampatibility andcould be used as a bone graft substitute inmaxillofacial plastic repair. This kind of material could be carved and contoured ...
Objective To investigate the relationship between the initial stabil ity and infected loosening of the total hip arthroplasty (THA) prosthesis. Methods From January 2000 to December 2008, 110 cases (110 hips) were treated with THA revision. Among them, 15 cases (15 hips) were confirmed infected loosening. There were 8 males and 7 females with anaverage age of 62 years (range 42-75 years). The infected signs were found from 6 months to 2 years after initial THA. All of them had Tsukayama type IV and late infection, including 6 cases of acetabular infected loosening (5 cases of one-stage and 1 case of two-stage acetabular revision), 7 cases of simple infected loosening of femoral prosthesis (4 cases of one-stage and 3 cases of twostage femoral prosthesis revision), and 2 cases of joint capsule infection and sinus without prosthesis loosening (debridement and continuous irrigation). Results All incisions healed by first intention. Fifteen patients were followed up for 12 to 36 months (average 24 months). In 13 cases of revision, postoperative X-ray films showed that femoral acetabular prostheses were in good position, and had no cl inical and imaging infective signs of loosening. In 2 cases of joint capsule infection, sinus recurred 6 months postoperation without hip joint pain, the function of weight-bearing and walking of hip joint was normal. Harris score increased from preoperative average of 42 to postoperative average of 85; the results were excellent in 4 cases, good in 7 cases, and fair in 4 cases. Conclusion The infection of THA may occur in the whole joint, half-joint or just in joint capsule. The initial stabil ity of the prosthesis would affect the long-term survival of the prothesis. If the prosthesis initial stabil ity is obtained, even if there are infective factors, infections would also be l imited.
Objective To evaluate the effectiveness of limited open reduction and double plates internal fixation in the treatment of Vancouver type B1 periprosthetic femoral fracture after hip arthroplasty. Methods A retrospective analysis was made on the clinical data of 12 patients with Vancouver type B1 periprosthetic femoral fracture after hip arthroplasty between May 2007 and October 2012, who underwent limited open reduction and double plates internal fixation. Of 12 patients, 4 were male and 8 were female, aged 76-85 years (mean, 81.6 years); the left side was involved in 5 patients and the right side in 7 patients. The cement prosthesis was used in 3 cases and cementless prosthesis in 9 cases; double acting head of hip arthroplasty was performed in 4 cases and total hip arthroplasty in 8 cases. The median time from first hip arthroplasty to re-fracture was 13 months (range, 5 months-5 years). The causes of injury were traffic accident in 2 cases and falling in 10 cases. Combined fractures included contralateral tibial and fibular fractures and ipsilateral distal humeral fracture (1 case), ipsilateral proximal humeral fracture (2 cases), ipsilateral distal radial fracture (1 case), and rib fracture (1 case). The time from injury to operation was 5.6 days on average (range, 3-10 days). Results The incisions all healed by first intention, and no infection or deep venous thrombosis of lower extremity occurred. Twelve cases were followed up 6-24 months (mean, 13.3 months). One female patient died of acute myocardial infarction at 16 months after operation. All the fractures were healed, with X-ray healing time of 12.5 weeks on average (range, 10-16 weeks). The time of full weight bearing was 13 weeks on average (range, 10-18 weeks). Ten cases could walk freely after operation, and 2 cases could walk by the aid of walking aid appliance. At last follow-up, the Harris score of hip function was 87.3 on average (range, 75-93). Conclusion The method of limited open reduction and double plates internal fixation for Vancouver type B1 periprosthetic femoral fracture after hip arthroplasty is effective in maintaining stability, protecting blood supply, promoting fracture healing, and doing functional exercise early. The long-term effectiveness needs further observation.
目的:總結空心螺釘、DHS或人工關節置換治療股骨頸骨折的適應證及效果。方法:分別用空心螺釘、DHS固定和人工關節置換3種方法治療獲得隨訪的中老年股骨頸骨折29例。年齡51～85歲,平均68.3歲。空心螺釘組8例,DHS組15例,人工關節組6例。結果:隨訪29例,隨訪時間1～3年,平均2~3年。圍手術期無1例死亡,無切口感染,無內固定斷裂,無關節脫位。發生退釘2例,股骨頭壞死1例。功能評定:優20例,良6例,差3例。結論:根據患者的年齡、經濟狀況、骨折部位、骨折分型合理地選擇治療方法,是減少并發癥,提高治療效果和老年人生活質量的關鍵。
ObjectiveTo investigate the efficacy of total hip arthroplasty (THA) assisted by the MAKO robotic arm via posterolateral approach.MethodsThe clinical data of 70 patients treated with THA via posterolateral approach between March 2017 and March 2019 who met the selection criteria were retrospectively analyzed. According to different treatment methods, the patients were divided into two groups, 35 were treated with MAKO robotic arm assisted THA (MAKO group) and 35 with traditional THA (THA group). There was no significant difference in gender, age, body mass index, disease duration, etiology, perioperative time, preoperative activity of daily living (ADL) scale index, American Society of Anesthesiologists (ASA) classification, walking ability, comorbidities, hemoglobin, and other general data between the two groups (P>0.05). The operation time, intraoperative blood loss, hospital stay, postoperative acetabular abduction and anteversion angles, postoperative length difference of bilateral lower limbs, and proportions of intraoperative blood transfusion, immediate postoperative loading, wound drainage time more than 2 days, and complications were recorded and compared between the two groups. According to the X-ray films at 6 months after operation, the reduction quality was judged. The forgotten joint score, Harris score, and proportions of independent walking and ADL index increased were used to evaluate the function recovery of patients.ResultsPatients in both groups were followed up 6-18 months, with an average of 8 months. There was no significant difference (P>0.05) between the two groups in operation time, intraoperative blood loss, hospital stay, acetabular abduction and anteversion angles, and length difference of both lower limbs at 6 months after operation. There was no significant difference in the proportions of intraoperative blood transfusion, immediate postoperative loading, and wound drainage time more than 2 days between the two group (P>0.05). X-ray reexamination at 6 months after operation showed that there was no significant difference in the reduction quality between the two groups (Z=4.191, P=0.123). Postoperative complications occurred in 7 patients (20.0%) in the MAKO group and 10 patients (28.6%) in the THA group, showing no significant difference in the incidence of complications between the two groups (χ2=2.121, P=0.224). Two patients (5.7%) in the MAKO group and 4 patients (11.4%) in the THA group underwent revision within 6 months, showing no significant difference in the revision rate between the two groups (χ2=0.729, P=0.673). At 3 and 6 months after operation, the proportions of independent walking and ADL index increased showed no significant difference between the two groups (P>0.05). Harris scores in both groups improved significantly when compared with preoperative scores (P<0.05); there was no significant difference in the forgotten joint scores and Harris scores between the two groups (P>0.05).ConclusionCompared with traditional THA, MAKO robotic arm assisted THA has longer operation time and more intraoperative blood loss, but it has the advantages of accurate positioning and simple operation, and there is no significant difference in short-term postoperative function recovery.
ObjectiveTo apply the multi-modal deep learning model to automatically classify the ultra-widefield fluorescein angiography (UWFA) images of diabetic retinopathy (DR). MethodsA retrospective study. From 2015 to 2020, 798 images of 297 DR patients with 399 eyes who were admitted to Eye Center of Renmin Hospital of Wuhan University and were examined by UWFA were used as the training set and test set of the model. Among them, 119, 171, and 109 eyes had no retinopathy, non-proliferative DR (NPDR), and proliferative DR (PDR), respectively. Localization and assessment of fluorescein leakage and non-perfusion regions in early and late orthotopic images of UWFA in DR-affected eyes by jointly optimizing CycleGAN and a convolutional neural network (CNN) classifier, an image-level supervised deep learning model. The abnormal images with lesions were converted into normal images with lesions removed using the improved CycleGAN, and the difference images containing the lesion areas were obtained; the difference images were classified by the CNN classifier to obtain the prediction results. A five-fold cross-test was used to evaluate the classification accuracy of the model. Quantitative analysis of the marker area displayed by the differential images was performed to observe the correlation between the ischemia index and leakage index and the severity of DR. ResultsThe generated fake normal image basically removed all the lesion areas while retaining the normal vascular structure; the difference images intuitively revealed the distribution of biomarkers; the heat icon showed the leakage area, and the location was basically the same as the lesion area in the original image. The results of the five-fold cross-check showed that the average classification accuracy of the model was 0.983. Further quantitative analysis of the marker area showed that the ischemia index and leakage index were significantly positively correlated with the severity of DR (β=6.088, 10.850; P＜0.001). ConclusionThe constructed multimodal joint optimization model can accurately classify NPDR and PDR and precisely locate potential biomarkers.