RAIQC

🎉 We are proud to share more results from the RAIQC’s study on AI-assisted interpretation for Chest X-ray (CXR), which has won the trainee prize at RSNA 2024! 🏆Congratulations to the RAIQC and the #Oxford team for this achievement. This multi-reader, multi-case study highlights the impact of Lunit INSIGHT CXR, showcasing its value across a variety of patient-facing specialties and experience levels. The study assessed 30 clinicians - including radiologists, reporting radiographers, and emergency, general internal, and intensive care physicians - each tasked with identifying 10 pathologies across 500 CXRs. With Lunit INSIGHT CXR, all specialties achieved statistically significant performance improvements in detecting 8 out of 10 targeted pathologies. 💡 This shows how AI assistance for CXR reading can empower clinicians of all experience levels to make better, faster decisions in high-stakes environments, especially as hospitals face increasing case volumes and complexity. Scroll through the carousel for more details! ☑️ Read more about it on the RAIQC website : https://bit.ly/4gq1IW0 #lunit #raiqc #rsna #study #publication #cxr #lunitinsight #ED #ER #chest #xray #radiology #clinician

This Lung Cancer Awareness Month, we want to talk to you about a dedicated nurse-led service at OUH that has been setting an example by delivering safe, effective and highly efficient patient-centred care, as evidenced by their consistently excellent performance over the last two years. The Lung Nodule service comprises two specialist nurses and an administrator. The team works alongside respiratory physicians and thoracic radiologists to analyse CT images and determine whether patients can be discharged, or need follow-up scans or referrals. The team receives an average of 80 referrals a month via OUH clinicians and GPs. One of the main roles of the nurses upon receiving a referral is to analyse CT images for any solid nodules and determine whether the patient can be discharged, or if they need follow up scans or a referral to the lung cancer team. The nurses discuss complex cases, possible cancers, and growing nodules with the respiratory and radiology consultants and make further action plans as needed. The service is largely virtual, but there is clinic time twice a week to see patients face-to-face and discuss more complex cases. Initially having to tackle a six-month backlog – which they cleared within months – the service has been noted for impressive delivery and development, with a service template for other similar centres across the country to use. This will benefit patients and ensure consistency of service, with an improvement in lung cancer detections rates and cost savings. This programme has shown improvements across the service, with: - A reduction in time from CT scanning to letting patients know their results - A reduction in patients needing follow-up scans - A projected significant increase in savings for OUH. 🗣️ Rachel Benamore, Lead Consultant Radiologist for the service said: "I'm extremely proud of the service that we have built with the nodule navigators, who provide safe and effective care for our patients. The service has proved to be immensely helpful in reducing waiting time for patients, which is really important at what can be quite a worrying time for them." 🗣️Professor Andrew Brent, Chief Medical Officer at OUH, said: "It is important and appreciated that our teams find innovative solutions to make our services efficient and reduce patient waiting times. 🗣️ "We are thrilled to see the positive impact of the lung nodule service, and are excited to see further improvements across it." You can read more about the service in our news story ⬇️ 🔗 https://lnkd.in/eQtTCKnf

💡 Explore Our Groundbreaking Abstracts at #RSNA2024! We’re excited to unveil #20 #abstracts showcasing our advanced AI solutions for chest and breast screening! 🏆 The RAIQC study on AI-assisted Chest X-ray (CXR) interpretation has received the RSNA 2024 Trainee Prize! Congratulations to RAIQC and the Oxford team for this outstanding achievement. Don't miss a special presentation on S:t Görans Hospital’s (Capio Sverige) year-long experience using AI to replace one radiologist with transformative results in their double-reading breast screening process. 👉 Check out all 20 abstracts and book a meeting with us to learn more: https://bit.ly/4egecOs #lunit #volpara #rsna2024 #rsna #abstract #radiology #radiologist #cancerscreening #cancer #chestscreening #breastscreening #breastcancer #study

This study investigates the effect of AI-assisted image interpretation on clinicians' ability to diagnose pneumothoraces using chest X-rays. 1️⃣ AI assistance increased overall sensitivity by 11.4%, from 66.8% to 78.1%. 2️⃣ Specificity slightly increased from 93.9% to 95.8%, though not statistically significant. 3️⃣ The largest sensitivity improvement (21.7%) was observed among junior clinicians. 4️⃣ Diagnostic confidence improved, particularly for junior clinicians, aligning their performance with more experienced peers. 5️⃣ Interpretation time remained consistent with and without AI assistance. 6️⃣ The study used a multicenter, multi-reader approach with 18 clinicians from various specialties and experience levels. This paper is worth reading for its comprehensive analysis of how AI tools can enhance diagnostic accuracy and confidence, particularly benefiting less experienced clinicians in acute care settings. ✍🏻 Alex Novak, Sarim Ather, Avneet Gill, Peter Aylward, Giles Maskell, Gordon W Cowell, Abdala Trinidad Espinosa Morgado, Tom Duggan, Melissa Keevil, Oliva Gordon, Osama Akrama, Elizabeth Belcher, Rhona Taberham, Rob Hallifax, Jasdeep Bahra, Abhishek Banerji, Jon Bailey, Antonia James, Ali Ansaripour, Nathan Spence, John Wrightson, Waqas Jarral, Steven Barry, Saher Bhatti, Kerry Astley, Amied Shadmaan, Sharon Ghelman, Alec Baenen, Jason Oke, Claire Bloomfield, Mark Beggs, Fergus Gleeson. Evaluation of the impact of artificial intelligence-assisted image interpretation on the diagnostic performance of clinicians in identifying pneumothoraces on plain chest X-ray: a multi-case multi-reader study. Emerg Med J. 2024. DOI: 10.1136/emermed-2023-213620