50 Trending Radiology Thesis Topics for MD/DNB (2025): AI, Interventional, and Nuclear Medicine

Introduction

Choosing the right thesis topic is arguably the most critical decision of your radiology postgraduate career. It’s a high-stakes choice that shapes your sub-specialty focus, dictates your time investment, and defines your publication record.

The reality? Too many postgraduates select topics that are too broad, lack good data, or simply won’t drive your career forward.

To help you overcome this hurdle, the Radiobite team—curated by senior residents and published radiologists—has compiled a list of 50 Trending Radiology Thesis Topics for 2024. We’ve focused on the three highest-impact areas shaping modern radiology: Artificial Intelligence (AI), Interventional Radiology (IR), and Nuclear Medicine (NM).

These topics offer a perfect blend of high-impact research potential and practical feasibility for the modern MD/DNB curriculum.


1. The Future is Now: AI & Radiomics Thesis Topics (n=20)

Artificial Intelligence is no longer a concept—it’s a clinical tool. Thesis work in this area is highly sought after and offers the best chance for high-tier publications.

AI in Image Interpretation & Workflow

  1. AI-assisted detection of incidental pulmonary nodules on routine CT: A comparison of performance metrics with junior residents.
  2. Evaluation of deep learning models for automated classification of liver lesions on contrast-enhanced MRI.
  3. The role of AI in reducing false positives and unnecessary recalls in a breast cancer screening program.
  4. Developing and validating an AI model for automated bone age assessment from hand X-rays in a local pediatric population.
  5. Assessing the efficiency of AI-powered tools in prioritizing urgent CT scans in the emergency department workflow.

Radiomics and Prognosis

  1. Radiomics signatures on pre-treatment MRI to predict therapeutic response in rectal cancer after neoadjuvant chemoradiotherapy.
  2. Quantitative texture analysis (Radiomics) of adrenal lesions on unenhanced CT to differentiate benign from malignant.
  3. Developing a Radiomics model from CT data to predict severity and outcome in acute pancreatitis (Modified CT Severity Index comparison).
  4. MRI-based texture analysis for non-invasive grading of gliomas (WHO Grade II vs. Grade IV) with histopathological correlation.
  5. The utility of Radiomics in characterizing renal masses on multiphasic CT and predicting their surgical pathology.

Large Language Models (LLMs) in Radiology

  1. Evaluating the performance of ChatGPT (or other LLMs) for drafting radiology reports: Accuracy, completeness, and time saving analysis.
  2. LLM-assisted generation of structured reporting templates for complex oncologic follow-up studies.
  3. Assessing the use of LLMs in generating patient-friendly radiology reports and summaries.
  4. Using Generative AI to create synthetic medical images for resident training: A comparison of diagnostic confidence.
  5. The ethical and legal implications of integrating LLM-generated content into clinical radiology practice.

Feasibility and Technical Topics

  1. Analysis of artifacts and challenges in training AI models across different vendor platforms (multi-site data).
  2. A study on the correlation between AI-reported segmentation errors and low image quality/motion artifacts.
  3. Evaluating the clinical impact of incorporating AI into the existing PACS viewer for chest X-ray triaging.
  4. Inter-reader variability in radiologists using an AI-assisted detection tool for mammography.
  5. The role of AI in optimizing low-dose CT acquisition protocols without compromising diagnostic quality.

2. Interventional Radiology (IR) Thesis Topics (n=15)

IR topics are excellent for residents who want a mix of clinical data collection, procedural follow-up, and demonstrating clinical impact.

  1. Comparison of trans-arterial chemoembolization (TACE) vs. Yttrium-90 Radioembolization (TARE) outcomes for unresectable hepatocellular carcinoma.
  2. Assessing the technical success and clinical outcomes of percutaneous cryoablation vs. radiofrequency ablation for renal cell carcinoma ( <4 cm).
  3. Efficacy of vertebral augmentation (kyphoplasty/vertebroplasty) in reducing pain scores in malignant versus osteoporotic compression fractures.
  4. The role of pre-procedural contrast-enhanced CT in predicting success rates of prostatic artery embolization (PAE) for benign prostatic hyperplasia (BPH).
  5. Evaluation of various sclerosant agents in the treatment of post-traumatic and spontaneous lymphoceles and cystic lesions.
  6. Technical success and complication rates of percutaneous biliary drainage vs. endoscopic stenting in malignant biliary obstruction.
  7. A retrospective analysis of complication rates following image-guided core needle biopsy of thoracic and abdominal masses.
  8. Outcomes of uterine fibroid embolization (UFE): A comparison of symptomatic relief between different embolizing agents.
  9. Clinical outcomes and patient satisfaction after percutaneous drain removal in intra-abdominal abscesses.
  10. The safety and effectiveness of catheter-directed thrombolysis for sub-massive pulmonary embolism in the acute setting.
  11. Comparative study of transradial versus transfemoral approach for visceral angiography procedures.
  12. Bland embolization for splenic trauma: Assessing organ preservation and complication rates.
  13. Efficacy of endovascular treatment for chronic total occlusions (CTO) in peripheral arterial disease patients.
  14. The use of image-guided ablative techniques for pain palliation in musculoskeletal metastases.
  15. Pre-operative embolization of vascular tumors (e.g., juvenile nasopharyngeal angiofibroma): Technical challenges and blood loss reduction.

3. Nuclear Medicine (NM) / PET Thesis Topics (n=15)

As PET/CT and PET/MRI gain prominence, research in Nuclear Medicine offers a unique blend of molecular imaging and clinical management.

  1. Diagnostic utility of PSMA PET/CT in biochemical recurrence of prostate cancer: A comparison with conventional imaging.
  2. F-18 FDG PET/CT in the assessment of fever of unknown origin (FUO) and its impact on clinical management.
  3. Evaluation of response to treatment in lymphoma patients using the Deauville scoring system on PET/CT.
  4. The role of Lu-177 DOTATATE therapy (Theranostics) outcomes for somatostatin receptor-positive neuroendocrine tumors.
  5. Comparative study of DOTA-TOC/TATE PET versus FDG PET in the initial staging of neuroendocrine tumors.
  6. The incremental value of PET/MRI fusion imaging over PET/CT in staging and restaging head and neck cancers.
  7. Assessing the utility of F-18 FDG PET/CT in the detection and characterization of large vessel vasculitis.
  8. Correlation of myocardial perfusion imaging findings with coronary angiography in patients with suspected coronary artery disease.
  9. The impact of respiratory and cardiac gating on quantitative assessment of tumor response on PET/CT.
  10. Pre-surgical localization of hyper-functioning parathyroid adenomas using Tc-99m Sestamibi scan and SPECT/CT.
  11. Tc-99m labeled red blood cell scintigraphy for the detection of gastrointestinal bleeding: Diagnostic yield and localization.
  12. Comparative performance of PET radiotracers for imaging Alzheimer’s disease (e.g., Amyloid vs. Tau tracers).
  13. Assessing the incidence and causes of false-positive FDG uptake in non-malignant conditions (e.g., inflammation, infection).
  14. The role of DAT scans in differentiating essential tremor from Parkinson’s disease: Accuracy and clinical correlation.
  15. Evaluation of I131 therapy outcomes and long-term follow-up in patients with differentiated thyroid cancer.

Conclusion and Next Steps

Finding a great thesis topic is the first step toward a successful academic career. By focusing on cutting-edge areas like AI, IR, and NM, you position yourself at the forefront of the specialty.

Don’t let the planning process overwhelm you. A solid proposal and timeline are essential for timely completion.

Comments

No comments yet. Why don’t you start the discussion?

Leave a Reply

Your email address will not be published. Required fields are marked *