An active effort in our group is on developing methodologies for task-specific imaging. This is a highly exciting area of research, which, at its core, is based on the idea of optimizing imaging algorithms based on their performance in specific clinical tasks. This leads to innovation in multiple ways. Here are a couple of youtube videos from an invited lecture given by the lab PI, Dr. Jha, that articulate this research direction. 

Fundamentals of task-specific evaluation
Task-specific evaluation of AI algorithms

In our most recent efforts, we have had the honor of leading multi-institutional efforts to lay out guidelines for evaluation of artificial intelligence-based algorithms, including the recently published RELAINCE guidelines and a framework for objective task-based evaluation of AI algorithms.

A specific area of interest in our lab is on developing techniques to optimize quantitative imaging methods using patient data. Such optimization can substantially accelerate the clinical translation of these methods but is complicated by the lack of ground truth. We are developing no-gold-standard evaluation techniques to address this issue. Our results demonstrate that these techniques, without access to any ground truth, are able to rank imaging methods based on how precisely they estimate the  true quantitative value. A video presentation that describes our latest innovation in this area is on this page.

We are also developing methods to compute observer performance on detection tasks. The video below describes one of these efforts.


  • A. K. Jha, T. J. Bradshaw, I. Buvat, M. Hatt, P. KC, C. Liu, N. F. Obuchowski, B. Saboury, P. J. Slomka, J. J. Sunderland, R. L. Wahl, Z. Yu, S. Zuehlsdorff, A. Rahmim, and R. Boellaard, “Nuclear medicine and artificial intelligence: Best practices for evaluation (the RELAINCE guidelines),”, J. Nuc. Med., 2022 (open access link) (Media coverage)
  • A. K. Jha, K. J. Myers, N. A. Obuchowski, Z. Liu, M. A. Rahman, B. Saboury, A. Rahmim, B. A. Siegel, “Objective task-based evaluation of artificial intelligence-based medical imaging methods: Framework, strategies and role of the physician”, PET Clinics (special issue on AI in PET), 2021 (link) (arxiv) (Media coverage)
  • Z. Liu, Z. Li, J. Mhlanga, B. Siegel, A. K. Jha, “No-gold-standard evaluation of quantitative imaging methods in the presence of correlated noise”, Proc. SPIE Med. Imag. 2022 (Finalist for Robert F. Wagner Best Student paper award) (link) (arxiv)
  • M. A. Rahman, Z. Yu, A. K. Jha, “Ideal-observer computation with anthropomorphic phantoms using Markov Chain Monte Carlo”, IEEE International Symp. Biomed. Imag. 2022 (link) (arxiv)
  • J. Liu, Z. LiuH. S. MoonJ. Mhlanga and A. K. Jha, “A no-gold-standard technique for objective evaluation of quantitative nuclear-medicine imaging methods in the presence of correlated noise”, J. Nucl. Med. May 2020, 61 (supplement 1) 523 (link)
  • A. K. Jha, E. Mena, B. Caffo, S. Ashrafinia, A. Rahmim, E. C. Frey and R. Subramaniam, “Practical no-gold-standard framework to evaluate quantitative imaging methods: Application to lesion segmentation in PET”, J. Med. Imag. (Special Section on PET imaging), 4(1), 2017 (pdf) (link).
  • A. K. Jha, B. Caffo, E. Frey, “A no-gold-standard technique for objective assessment of quantitative nuclear-medicine imaging methods”, Phys. Med. Biol., 61(7), 2780-2800, 2016 (pdf)
  • A. K. Jha, N. Song, B. Caffo, E. C. Frey,  Objective evaluation of reconstruction methods for quantitative SPECT imaging in the absence of ground truth. Proc SPIE Int Soc Opt Eng. 2015 Apr 13;9416:94161K.
  • A. K. Jha, M. A. Kupinski, J. J. Rodriguez, R. Stephen, A. Stopeck, “Task-based evaluation of segmentation algorithms for diffusion-weighted MRI without using a gold standard”, Phys. Med. Biol., 57(13) 4425-46, 2012 (link)
  • A. K. Jha, M. A. Kupinski, J. J. Rodriguez, R. M. Stephen, A. T. Stopeck. Evaluating segmentation algorithms for diffusion-weighted MR images: a task-based approach. Proc. SPIE Medical Imaging; San Diego, CA, USA. Feb 2010.pp. 76270L1–L8, Best Student Paper award (link)