ItemPre-op localisation of impalpable breast lesions with a radiofrequency identifier device (RFID)(2023-05) Ravichandran, Duraisamy; Zakharova, Natalia; Shrestha, Deepak; Murthy, Sreedhar; Butler, Stephanie; Swinson, Christine; Kirkpatrick, KatharineImpalpable breast lesions are traditionally localised with a wire to facilitate surgical removal. This method of localisation is associated with a number of logistical problems, the main one being the requirement to do the localisation on the day of surgery. More recently, a number of non-wire techniques have been introduced. We evaluated the performance of a RFID (LOCalizerTM) in this context. ItemBeware of Superficial Thrombophlebitis Mimicking a Foreign Body(2023) Georgiades, Fanourios; Najem, MojahidA 37 year old woman was admitted with diabetic ketoacidosis. She later complained of pain, swelling, and a rash at a previous venous catheter site. The patient was afebrile, with an erythematous, oedematous, tender right forearm without any fluctuant areas. Ultrasound identified an irregular 4 cm echogenic material (A) within the right cephalic vein, occupying part of the lumen that was partially compressible (B). This was reported as an intravenous foreign body and may have represented a retained cannula catheter. The vascular multidisciplinary team disagreed and concluded this was superficial thrombophlebitis. The patient was treated with non-steroidal anti-inflammatory drugs and recovered well. ItemRadiation associated angiosarcoma of the breast: a pictorial review of eleven cases highlighting a devastating complication of breast radiotherapy(2023) Stewart-Parker, Emma; Kirkpatrick, Katharine; D'Souza, Lorraine; Ravichandran, DuraisamyRadiation Associated Angiosarcoma (RAAS) of the breast is a rare secondary malignancy following breast radiotherapy and associated with considerable diagnostic and therapeutic challenges. We report the experience of one UK breast unit with a case series diagnosed over the last 22 years. Data was prospectively collected on all patients who presented with RAAS from 2001 to 2023, including photographs obtained following formal patient consent. The characteristics of the original breast carcinoma, treatment received, presenting features of the RAAS, subsequent management and patient outcomes were recorded. Eleven patients were identified, all women, with a median age of 57 (range 39–82) at initial breast cancer diagnosis. The dose of breast radiation received varied from 40 to 70 Gy (median 50), including boosts. The commonest presentation of RAAS was with recent-onset skin discolouration at a median time of 70 months (range: 53–168 months) after radiotherapy. Mammograms and ultrasounds were often negative. Diagnosis was confirmed largely by punch biopsy, and 82% (n = 9) went on to have surgery. The median survival from the diagnosis of RAAS was 13 months (range 7–101). RAAS is a devastating complication of breast radiotherapy, with poor prognosis. As it often presents after the traditional 5-year follow-up period for breast cancer, can appear similar to post-radiotherapy changes and be associated with “normal” imaging, so the diagnosis hinges almost entirely on clinical suspicion. Thus, it is important to be aware of the possibility of RAAS in post radiotherapy breast cancer patients. ItemAutomated Identification of Orthopedic Implants on Radiographs Using Deep Learning(2021-03-07) Batta, VineetAccurate identification of metallic orthopedic implant design is important for preoperative planning of revision arthroplasty. Surgical records of implant models are frequently unavailable. The aim of this study was to develop and evaluate a convolutional neural network for identifying orthopedic implant models using radiographs. In this retrospective study, 427 knee and 922 hip unilateral anteroposterior radiographs, including 12 implant models from 650 patients, were collated from an orthopedic center between March 2015 and November 2019 to develop classification networks. A total of 198 images paired with autogenerated image masks were used to develop a U-Net segmentation network to automatically zero-mask around the implants on the radiographs. Classification networks processing original radiographs, and two-channel conjoined original and zero-masked radiographs, were ensembled to provide a consensus prediction. Accuracies of five senior orthopedic specialists assisted by a reference radiographic gallery were compared with network accuracy using McNemar exact test. When evaluated on a balanced unseen dataset of 180 radiographs, the final network achieved a 98.9% accuracy (178 of 180) and 100% top-three accuracy (180 of 180). The network performed superiorly to all five specialists (76.1% [137 of 180] median accuracy and 85.6% [154 of 180] best accuracy; both P , .001), with robustness to scan quality variation and difficult to distinguish implants. A neural network model was developed that outperformed senior orthopedic specialists at identifying implant models on radiographs; real-world application can now be readily realized ItemKnee Implant Identification by Fine-Tuning Deep Learning Models(2021-09-28) Batta, Vineet; Kiruthika, M.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.