AJR Podcasts

A strong research question can still fail if the study design cannot answer it. Michael Soulen, MD, and Shudaveep Ganguly, MD, DM, speak with host Amit Gupta, MD, about selecting meaningful outcomes, matching design to the question, avoiding pitfalls of diagnostic accuracy studies, involving statisticians early, and building multidisciplinary trials. Listen to their discussion in episode 3 of The Early Career Researcher's Playbook, an AJR Podcast Series. *Key Takeaways Meaningful Outcomes vs. Surrogates: Technical success and progression-free survival are only surrogate endpoints. Patients and clinicians ultimately care about overall survival and quality of life. Imaging surrogates are notoriously poor predictors of true clinical benefit. Flaws in Diagnostic Accuracy: Relying purely on sensitivity and specificity is problematic if the study population doesn't reflect real-world conditions. Likelihood ratios may provide a realistic indication of a test's clinical impact. Early Statistical Integration: Never treat a statistician as a data calculator at the end of a study. Engaging them during the brainstorming phase ensures feasibility and helps to properly structures your protocol and database. *Chapters  0:00 - Welcome 2:23 - Choosing Meaningful Outcomes 9:28 - RCTs and Real World Limits 18:36 - Diagnostic Study Pitfalls 25:58 - From Accuracy to Impact 28:55 - Statistics Starts Early 37:45 - Making Collaboration Work 47:01 - Quick Fire Playbook 51:04 - Final Advice Follow AJR on Social Media LinkedIn: https://www.linkedin.com/showcase/ajr-radiology/ YouTube: https://www.youtube.com/channel/UCfFAYezkLMxJGMgIJLN0Dpg Instagram: https://www.instagram.com/ajr_radiology/ TikTok: https://www.tiktok.com/@ajr_radiology X: https://x.com/AJR_Radiology BlueSky: https://bsky.app/profile/ajrradiology.bsky.social Threads: https://www.threads.com/@ajr_radiology *These sections were generated using artificial intelligence (Descript and Google Gemini) and then reviewed for accuracy.

Good research questions may be hiding in everyday clinical work. Marco Francone, MD, PhD, and Daniel Pinto dos Santos, MD, speak with host Amit Gupta, MD, about transforming clinical intuition, recurring gaps, and diagnostic challenges into focused, testable, and clinically impactful research. Listen to their discussion in episode 2 of The Early-Career Researcher's Playbook, an AJR Podcast Series. https://www.ajronline.org/doi/10.2214/AJR.26.35769 *Key Takeaways Questioning the Routine: Meaningful research often begins by simply asking "why" established practices exist. For example, questioning the necessity of a non-contrast series prior to contrast media administration when looking for abdominal hemorrhage may lead to a publishable study.   The Prospective Shift: Retrospective research may be limited by biased intuitions and imperfect hypotheses. The specialty should move toward prospective models, carefully considering the clinical impact and cost-effectiveness of potential interventions.   The "Tool-First" Trap: Pushed by a publish-or-perish academic culture, researchers may make the mistake of starting with an AI model or radiomics pipeline and then searching for a problem to which to apply it. True clinical impact more likely comes with starting with the patient problem and finding the appropriate method to solve it.   Combating Confirmation Bias: When researchers firmly believe their own clinical intuition, they risk designing studies solely to confirm their preconceived notions. To maintain scientific credibility, you should assign a colleague the specific role of opposing your views to actively try and prove your hypothesis wrong.  *Chapters 0:00 - Introduction 2:07 - What Clinical Intuition Means 6:47 - Daniel's Routine to Research 9:23 - Ideas Hiding in Workflow 11:32 - From Observation to Study Plan 14:28 - Marco's Prospective CT Example 18:26 - Avoiding Confirmation Bias 23:20 - AI Tool First Trap 29:53 - Quick Fire Takeaways 33:34 - Final Advice and Closing Follow AJR on Social Media LinkedIn: https://www.linkedin.com/showcase/ajr-radiology/ YouTube: https://www.youtube.com/channel/UCfFAYezkLMxJGMgIJLN0Dpg Instagram: https://www.instagram.com/ajr_radiology/ TikTok: https://www.tiktok.com/@ajr_radiology X: https://x.com/AJR_Radiology BlueSky: https://bsky.app/profile/ajrradiology.bsky.social Threads: https://www.threads.com/@ajr_radiology *These sections were generated using artificial intelligence (Descript and Google Gemini) and then reviewed for accuracy.

A good research idea is not always a good research question. Bruno Hochhegger, MD, PhD, speaks with host Amit Gupta, MD, about choosing research questions that are clinically relevant, feasible, and worth pursuing in real-world settings. Listen to their discussion in episode 1 of The Early Career Researcher's Playbook, an AJR Podcast Series. Full article: https://www.ajronline.org/doi/10.2214/AJR.26.35586 *Key Takeaways Defining Clinical Relevance: A strong research question must answer a direct patient management issue, such as determining if a nodule is benign or malignant. The Feasibility and Sustainability Matrix: Moving from simple case reports to professional science requires navigating EMRs, securing IRB authorizations, and finding reliable grant funding. The AI Implementation Trap: While building artificial intelligence models on GitHub has become more accessible, adequately validating these tools for real-world clinical practice remains a massive, frequently underestimated hurdle. Multidisciplinary Research Networks: True feasibility requires nonmedical input; successful projects demand early feedback from IT departments, technologists, and referring surgeons to ensure workflows survive reality. *Key Moments 00:00 Intro and Welcome 01:18 The Importance of Small Steps: Lessons from a Failed MRI Lung Cancer Screening Trial. 04:30 Defining Clinical Relevance and the Crucial Role of the Physician-Researcher. 07:47 Assessing Feasibility: Navigating EMRs, IRB Approvals, and Data Access. 08:45 Research Sustainability: Securing Grants and Transitioning from Voluntary to Professional Science. 10:50 The AI Feasibility Trap: Why Validating Models is Harder Than Coding on GitHub. 16:08 Identifying the Key Indication: Solving Direct Patient Management Questions. 21:51 Beyond Mentorship: Building a Network Across IT, Technologists, and Referring Clinicians. 28:51 Establishing Niche Expertise: Strategic Advice for Radiology Residents and Fellows Follow AJR on Social Media LinkedIn: https://www.linkedin.com/showcase/ajr-radiology/ YouTube: https://www.youtube.com/channel/UCfFAYezkLMxJGMgIJLN0Dpg Instagram: https://www.instagram.com/ajr_radiology/ TikTok: https://www.tiktok.com/@ajr_radiology X: https://x.com/AJR_Radiology BlueSky: https://bsky.app/profile/ajrradiology.bsky.social Threads: https://www.threads.com/@ajr_radiology *These portions of the page were generated using artificial intelligence (Google Gemini) and then reviewed for accuracy.

What happens when the system goes down in a cyberattack? Ichiro Ikuta, MD, MMSc, speaks with cohosts Lindsey Negrete, MD, and Amy Maduram, MD about real-world vulnerabilities and consequences of cyberattacks in radiology. Listen to their discussion in episode 3 of Extreme Radiology, an AJR Podcast Series.  https://www.ajronline.org/doi/10.2214/AJR.26.35847 *Key Takeaways The Power Dynamics of Phishing: Hackers may impersonate department chairs or hospital CEOs using spoofed emails to pressure staff into handing over login credentials. Always verify suspicious requests via a direct phone call or in-person conversation. The Threat of Outside CDs: Discs containing outside imaging studies are major vectors for malware. To help prevent spreading a digital infection, outside CDs should be uploaded in isolated, dedicated environments rather than directly on standard clinical workstations. AI and Blockchain in Cybersecurity: While blockchain technology offers decentralized tracking to protect financial and patient data, artificial intelligence is a double-edged sword, equipping both cybersecurity defenders and malicious hackers with significantly more powerful tools. *Chapters 0:00 - Opening 2:07 - Why Cybersecurity Matters 4:15 - How Hospitals Get Hacked 8:31 - Spotting Phishing Tricks 11:24 - When Systems Go Down 14:40 - Legal Duties and Training 16:25 - Recovery and Backlog Reality 20:01 - AI and Blockchain Threats 22:39 - Can You Hack It? Game 27:32 - Radiology Targets and Defenses 30:05 - Wrapping Things Up Follow AJR on Social Media LinkedIn: https://www.linkedin.com/showcase/ajr-radiology/ YouTube: https://www.youtube.com/channel/UCfFAYezkLMxJGMgIJLN0Dpg Instagram: https://www.instagram.com/ajr_radiology/ TikTok: https://www.tiktok.com/@ajr_radiology X: https://x.com/AJR_Radiology BlueSky: https://bsky.app/profile/ajrradiology.bsky.social *These sections were generated using artificial intelligence (Descript and Google Gemini) and then reviewed for accuracy.

Curious about the future of radiology economics? Richard E. Heller III, MD, MBA speaks with co-hosts Sherry S. Wang, MBBS, and Surbhi Raichandani, MD, to discuss radiology economics, from the basics to the niche and what is up ahead. Listen to their discussion in episode 2 of Difficult Conversations, an AJR Podcast Series. Full article: https://www.ajronline.org/doi/10.2214/AJR.26.35690 *Key Takeaways The Efficiency Fallacy: CMS recently implemented a 2.5% efficiency reduction to work relative value units, assuming that technology makes reading scans faster. However, modern multiplanar CTs contain thousands of images which can slow down down interpretation times. The No Surprises Act and Ghost Rates: While designed to protect patients from unexpected out-of-network bills, poor implementation allows insurers to use "ghost rates," which are contracted rates for never-billed services, to artificially lower the Qualifying Payment Amount and force rate reductions on practices. Advocacy is Essential: Radiologists serve as the nexus of the hospital. If the specialty does not demonstrate its downstream value, such as avoiding unnecessary surgeries and shortening hospital stays, business-minded policymakers will continue to treat it as a high-cost expense center to be cut. *Chapters  0:00 - Welcome 1:12 - Future Outlook For Radiology 2:27 -  How Radiologists Generate Revenue 3:09 - RVUs Conversion Factor Basics 4:30 - MIPS And Payment Adjustments 5:59 - Revenue Versus Paycheck 7:22 - Medicare Fee Schedule Trends 8:53 - Efficiency Adjustment Debate 12:50 - How CMS Sets The Rules 14:24 - Comment Letters And Advocacy 17:16 - Medicare As System Benchmark 19:09 - Commercial Insurer Headaches 21:21 - Steerage And Pediatric Risks 22:57 - No Surprises Act Explained 27:59 - Arbitration And Implementation Issues 29:10 - Radiology as Nexus 30:16 - Proving Downstream Value 31:34 - Total Value Equation 33:44 - RVUs and Complexity 36:53 - Fixing Incentives 46:04 - AI as Capacity Booster 49:01 - Will AI Cut Pay 50:12 - Teleradiology Done Right 54:30 - Radiology in 10 Years 57:53 - Optimism and Call to Action Follow AJR on Social Media LinkedIn: https://www.linkedin.com/showcase/ajr-radiology/ YouTube: https://www.youtube.com/channel/UCfFAYezkLMxJGMgIJLN0Dpg Instagram: https://www.instagram.com/ajr_radiology/ TikTok: https://www.tiktok.com/@ajr_radiology X: https://x.com/AJR_Radiology BlueSky: https://bsky.app/profile/ajrradiology.bsky.social Threads: https://www.threads.com/@ajr_radiology *These sections were generated using artificial intelligence (Descript and Google Gemini) and then reviewed for accuracy.