An evaluation of AI generated literature reviews in musculoskeletal radiology.

PURPOSE: The use of artificial intelligence (AI) tools to aid in summarizing information in medicine and research has recently garnered a huge amount of interest. While tools such as ChatGPT produce convincing and naturally sounding output, the answers are sometimes incorrect. Some of these drawbacks, it is hoped, can be avoided by using programmes trained for a more specific scope. In this study we compared the performance of a new AI tool (the-literature.com) to the latest version OpenAI's ChatGPT (GPT-4) in summarizing topics that the authors have significantly contributed to. METHODS: The AI tools were asked to produce a literature review on 7 topics. These were selected based on the research topics that the authors were intimately familiar with and have contributed to through their own publications. The output produced by the AI tools were graded on a 1-5 Likert scale for accuracy, comprehensiveness, and relevance by two fellowship trained consultant radiologists. RESULTS: The-literature.com produced 3 excellent summaries, 3 very poor summaries not relevant to the prompt, and one summary, which was relevant but did not include all relevant papers. All of the summaries produced by GPT-4 were relevant, but fewer relevant papers were identified. The average Likert rating was for the-literature was 2.88 and 3.86 for GPT-4. There was good agreement between the ratings of both radiologists (ICC = 0.883). CONCLUSION: Summaries produced by AI in its current state require careful human validation. GPT-4 on average provides higher quality summaries. Neither tool can reliably identify all relevant publications.

Outcomes Following Pedestal Cup Reconstruction of (Impending) Pathological Fractures of the Acetabulum due to Metastatic Bone Disease.

Management of periacetabular metastatic bone disease (MBD) is challenging, specifically if associated with bone loss or fracture. The aim of this study was to evaluate the complications and outcomes after undergoing peri-acetabular reconstruction using an 'ice-cream cone' pedestal cup endoprostheses for the most severe cases of (impending) pathological acetabular fractures. Fifty cases with severe periacetabular disease were identified. Acetabular defects were classified using the Metastatic Acetabular Classification (MAC). Pre- and post-operative mobility was assessed using the Eastern Cooperative Oncology Group (ECOG) Performance Status. Pain levels were assessed using a verbal rating scale. Surgical complications and patient survival were analysed; the Prognostic Immune Nutritional Index (PINI) was applied retrospectively to survival. There were 32 females and 18 males with a median age of 65 (41-88). Median post-operative follow-up was 16 months (IQR 5.5-28.5 months). Thirty-nine had complete, and 11, impending pathological fractures. The observed five-year survival was 19%, with a median survival of 16 months (IQR 5.8-42.5 months). Significantly worse survival was observed with PINI scores < 3.0 (p = 0.003). Excluding three perioperative deaths, 13 complications occurred in 12 patients: Implant failure in six patients (four aseptic loosening, one dislocation and one infection). At the final follow-up, mobility and pain levels were improved in 85% and 100%, respectively. Reconstruction of significant pelvic MBD with the 'ice-cream cone' reduces pain and improves mobility. Whilst the mortality rate is high, it remains a reasonable option for bed-bound, immobile patients. We advocate the use of an 'ice-cream cone' prosthesis for selected patients balancing the reported risks with the observed benefits. Supplementary Information: The online version contains supplementary material available at 10.1007/s13193-024-01917-x.