Search

BACKGROUND: Identification of implant model from primary knee arthroplasty in pre-op planning of revision surgery is a challenging task with added delay. The direct impact of this inability to identify the implants in time leads to the increase in complexity in surgery. Deep learning in the medical field for diagnosis has shown promising results in getting better with every iteration. This study aims to find an optimal solution for the problem of identification of make and model of knee arthroplasty prosthesis using automated deep learning models. METHODS: Deep learning algorithms were used to classify knee arthroplasty implant models. The training, validation and test comprised of 1078 radiographs with a total of 6 knee arthroplasty implant models with anterior–posterior (AP) and lateral views. The performance of the model was calculated using accuracy, sensitivity, and area under the receiver-operating characteristic curve (AUC), which were compared against multiple models trained for comparative in-depth analysis with saliency maps for visualization. RESULTS: After training for a total of 30 epochs on all 6 models, the model performing the best obtained an accuracy of 96.38%, the sensitivity of 97.2% and AUC of 0.985 on an external testing dataset consisting of 162 radiographs. The best performing model correctly and uniquely identified the implants which could be visualized using saliency maps. CONCLUSION: Deep learning models can be used to differentiate between 6 knee arthroplasty implant models. Saliency maps give us a better understanding of which regions the model is focusing on while predicting the results.

BACKGROUND: With increasing numbers of primary total hip replacement (THR), there has been a substantial increase in revision total hip replacement (RTHR) surgeries. RTHR are complex joint reconstruction surgeries involving significant cost, expertise and infrastructure. With its significant socioeconomic impact, we need to keep a close watch on the epidemiological trends of these procedures. METHODS: We prospectively studied the first-time RTHR performed at our institution for a 7-year period (2011–2017). We looked at patient demographics, the workload of RTHR and its etiology. We reviewed the microbiological profiles of septic revisions. RESULTS: Of the 1244 THR procedures performed, 260 (21%) were first-time revisions. The predominant cause of revisions was a prosthetic infection (38%) followed by aseptic loosening (33%), instability (15%), peri-prosthetic fracture (11%) and implant breakage (3%). In the aseptic loosening group, 55% of cases had primary cemented implant, 44% had only stem loosening, 31% had cup loosening and 25% had both cup and stem loosening. In the early, midterm, and late-failure groups, prosthetic infection remained the main cause of failure. In 60% of the septic revisions, the offending organisms could not be identified and of those identified most (77%) were gram negative. CONCLUSION: In our study, the RTHR burden was 21%, which is similar to historic revision data from the west (1998–2001) and twice as compared to recent trends from the west (9–11%). Unlike western data, which show aseptic loosening (30–60%) as the predominant cause of hip revisions, in our study infection was the number one cause (38%).

Background: There has been a gradual increase in the revision TKA (RTKA) workload due to expanding indications of total knee arthroplasty (TKA), coupled with improving patient longevity. Western countries are already looking at their data on RTKA to plan for the future heath care needs of these patients. Limited data is available on RTKA from developing countries. Our study attempts to fill this gap in knowledge. Materials and Methods: We prospectively documented details of all RTKA performed at our centre for a period of six years (2011-16). We recorded the volume, causes and time to failure from index surgery of all RTKA and further recorded microbiological pattern in septic failures. We looked at the proportion of each cause of failure and time from index surgery. Results: Of the 5068 TKA procedures performed from January 2011 to December 2016, 201 (4%) were first-time revisions. The predominant cause of revisions was prosthetic infection (61%) followed by aseptic loosening (18%) and instability (7%). In the early, mid term, and late-failure groups, prosthetic infection remained the main cause of failure. In 47% of the septic revisions, the offending organisms could be identified and of those identified most (67%) were Gram-negative. Conclusion: The volume of first-time RTKA procedures (4%) at our center remained low compared with that of the Western countries. In Western countries, the incidence of late aseptic failures was higher than that of early-septic failures, whereas in our study, revisions were more commonly performed in the early-failure group (48%) and most failures were due to prosthetic infection (61%).