Background/Objectives: Medical research institutions are increasingly leveraging artificial intelligence (AI) to enhance the processing and analysis of medical imaging data. However, scaling AI-driven medical image analysis often requires specialized expertise and infrastructure that individual labs may lack. A centralized solution is to establish a core facility—a shared institutional resource—dedicated to Automated Medical Image Processing and Analysis (AMIPA). Methods: This technical note offers a practical roadmap for institutions to create an AI-based core facility for AMIPA, drawing on our experience in building such a resource. Results: We outline the key components for replicating a successful AMIPA core facility, including high-performance computing resources, robust AI software pipelines, data management strategies, and dedicated support personnel. Emphasis is placed on workflow automation and reproducibility, ensuring researchers can efficiently and consistently process large imaging datasets. Conclusions: By following this roadmap, institutions can accelerate AI adoption in imaging workflows and foster a shared resource that enhances the quality and productivity of medical imaging research.
Background: Brain metastases (BMs) patients often receive multiple courses of radiotherapy (RT). Comparing follow-up imaging with RT plans is time-intensive and currently not possible in picture archiving and communication system (PACS). In this study, we aim to develop an automated tool to overlay pre-RT tumor contours onto follow-up MRI in the PACS to support response assessment and identification of untreated metastases, reducing review complexity and time. Methods: We built an AI-driven workflow that registers planning CT to follow-up MRI and propagates treated-lesion contours; outputs are exported as PACS-compliant DICOM. Performance was evaluated in 40 patients: 20 underwent quantitative comparison between manual and automated registrations using Dice similarity coefficient (DSC) and mean surface distance (MSD), following AAPM TG-132 guidelines; 20 additional patients (5-35 lesions per patient) underwent clinical evaluation of follow-up and pre-RT lesions contours visualization in PACS. Three physicians assessed these cases to measure review time and inter-observer agreement in treatment response classification (improved/stable vs indeterminate). Results: The workflow successfully registered all CT-MRI pairs. Mean DSC/MSD was: brain 0.97 ± 0.01/0.0 ± 0.0 mm, brainstem 0.89 ± 0.03/0.1 ± 0.1 mm, and gross tumor volumes 0.65 ± 0.18/0.6 ± 0.4 mm. Average physician review time per case decreased from 7.97 to 3.95 min with the automated workflow, and full inter-physician agreement increased from 72.4% to 93.5%. Conclusion: We developed and validated an AI-based tool that accurately fuses pre-RT contours with follow-up MRI for BMs, addressing a key gap in current PACS systems. The workflow significantly reduced review time per lesion and enhanced inter-physician agreement, and has the potential to enable faster, more consistent multidisciplinary follow-up assessment.
Background: Treatment of locally‐advanced head and neck cancer (HNC) involves multiple therapeutic exposures that may directly or indirectly impact cognition. Purpose: To describe long‐term cognitive impairment in a cohort of HNC survivors and to identify the cognitive domains most commonly affected. Methods and Materials: This study enrolled patients with locally advanced or recurrent HNC who received radiation therapy (RT) and survived at least 2 years without recurrence. Participants completed a cognitive battery and quality of life measures. Performance in a cognitive domain was considered impaired if T < 35. Logistic regression was used to assess the relationship between treatment exposures and other variables with impairment in each cognitive domain. Results: The final cohort consisted of 185 participants. The proportion of patients classified with cognitive impairment by domain was 36.8% for fine motor, 32.7% for verbal memory, 27.0% for executive function, 7.6% for vocabulary, and 7.0% for working memory. In the final model for impaired verbal memory, only radiation dose to the medial temporal lobe (OR = 1.06, 95% CI: 1.00–1.12) was significant. Time from RT (OR = 1.12, 95% CI: 1.01–1.24) and fatigue (OR = 3.45, 95% CI: 1.02–11.59) were associated with impaired working memory. Non‐white race (OR = 8.44, 95% CI: 3.23–22.05) and fatigue (OR = 2.68, 95% CI: 1.24–5.81) were associated with impaired executive function. Impaired emotional function and fatigue, but not cognitive battery performance, were associated with perceived cognitive impairment. Conclusions: This study supports the hypothesis that many long‐term HNC survivors experience cognitive impairment. Higher radiation dose to the medial temporal lobe was associated with lower verbal memory performance, and fatigue was associated with impaired executive function, perceived cognitive abilities, and overall role functioning.