New publication on quantitative SPECT for alpha-particle therapy

Delighted to share that our publication on a projection-domain low-count quantitative SPECT method for alpha-particle-based radiopharmaceutical therapies has been accepted for publication to the IEEE Transactions on Radiation Imaging and Plasma Sciences! The arxiv version of the article is available here. In this manuscript, we develop a method to perform quantification of uptake within different […]

Yan Liu awarded the SNMMI Student Research Grant

Yan Liu has been awarded the Society of Nuclear Medicine and Molecular Imaging (SNMMI) Student Research Grant! In this project, Yan will be developing new methods to quantify and harmonize PET radiomic features. Thanks to SNMMI for supporting this effort of Yan.

Eleven presentations at the 2022 SNMMI Annual Meeting

Excited to share that eleven abstracts based on work done by our lab and with collaborators have been accepted at the 2022 Society of Nuclear Medicine and Molecular Imaging (SNMMI) Annual Meeting to be held in Vancouver, Canada. One of these abstracts was first-authored by Yan Liu, who accomplished this research during her rotation period […]

Dr. Jha presents at the SNMMI AI Summit

The Society of Nuclear Medicine and Molecular Imaging (SNMMI) organized an AI summit with the goal of convening and gather various representatives from major stakeholders in nuclear medicine and AI spaces to identify and address barriers to the development of AI and ML tools for nuclear medicine, molecular imaging, and radiopharmaceutical therapy. Dr. Jha presented […]

Ziping finalist for Robert F. Wagner Best Student Paper Award at SPIE

Delighted to share that Ziping Liu was one of eight finalists for the Robert F. Wagner Best Student Paper Award at the SPIE Medical Imaging meeting. This was for his paper on no-gold-standard evaluation of quantitative imaging methods in the presence of correlated noise. Well done, Ziping!

Presentations at 2022 SPIE Medical Imaging

Zitong and Ziping presented talks at the SPIE Medical Imaging meeting on their works on developing an observer-study-based characterization and on no-gold-standard evaluation at the 2022 SPIE Medical Imaging meeting. Great job Zitong and Ziping!

Publication in Nature Medicine

Excited to share news of our new publication in Nature Medicine! AI algorithms often provide an uncertainty output, and we asked the question of how should this uncertainty be handled in clinical settings, leading to some fascinating ideas and findings! Give it a read here. Enjoyed the collaboration with Drs. Jonathan Birch (London School of […]

Yan Liu joins lab

Yan Liu, PhD student in the Imaging Sciences program, is the latest addition to our lab! Within her two-month rotation period, Yan generated and submitted an abstract on PET radiomics for the SNMMI 2022 Annual Meeting! Great job Yan and we are excited to have you join us!

Awarded NIH R56 grant

Excited to share that we have been awarded an NIH R56 grant on the topic of developing methods for oncological PET segmentation for the clinical goal of personalizing therapy in patients with non-small cell lung cancer. The Washington University media office did a press release on us receiving this grant. Link to press release here. […]

Publication on task-based evaluation of AI methods

Our article outlining a framework, strategies, and role of physicians in task-based evaluation of AI methods for medical imaging has been published in special issue on AI in PET Clinics! Excited about this highly collaborative multi-institutional effort! Link to article here and arxiv version here. The WashU media office did a press release on this […]

Invited talk at WMIC 2021 Annual Meeting

Dr. Jha gave an invited talk at the World Molecular Imaging Congress 2021 Annual Meeting on how  machine learning is pushing the boundaries of quantitative PET. The talk was followed by a stimulating panel discussion that centered on emerging areas in applying machine learning to medical imaging, challenges to address to make AI trustworthy, and […]

Ziping passes dissertation proposal exam

Ziping passed his thesis proposal exam with flying colors! His thesis proposal focused on developing novel image-analysis methods for nuclear-medicine imaging. Ziping’s dissertation committee consists of Dr. Clifford Robinson, Dr. Yuan Chuan Tai, Dr. Joyce Mhlanga, Dr. Deshan Yang and Dr. Jha. Many thanks to all the committee members for all their inputs during the […]

Alpha-SPECT work at Fully 3D meeting: Best oral award nominee

Excited to share that our own Zekun Li will be presenting our work on developing a multi-energy window quantitative SPECT method for joint quantification of Thorium and Radium! This work is a candidate for the best oral award nominee!  

Pre-print: Quantitative SPECT method for alpha-particle therapies

We posted a new pre-print that proposes and evaluates a projection-domain low-count quantitative SPECT method for alpha-particle emitting radiopharmaceutical therapy. The proposed method yields accurate significantly improved performance compared to state-of-the-art  methods for quantitative SPECT. Link here.

New publication on estimation-based PET segmentation

Our article on an estimation-based approach to segmenting PET images has been published in the Special issue on Early Career Researchers in Physics and Medicine and Biology! This article approaches a new approach to segmentation that accounts for the partial-volume effects, including the tissue-fraction effects in PET. Link to article here: A press release […]

Three presentations at SNMMI 2021 Annual Meeting

Three presentations by our team and collaborators are being presented at the SNMMI Annual Meeting 2021. Links below: Z. Li, N. Benabdallah, D. Abou, B. Baumann, R. Wahl, D. Thorek, A. K. Jha, “A projection-domain quantification method for absolute quantification with low-count SPECT for alpha-particle radiopharmaceutical therapy”, J. Nucl. Med. May 2021, 62 (supplement 1) 1539, Candidate for Best Poster […]

Awarded first R01!

We have been awarded our first R01! The grant is on the topic of no-gold-standard evaluation of quantitative imaging methods. More details here.

Three talks at SPIE Medical Imaging

We are excited to be presenting three talks at the upcoming SPIE Medical Imaging meeting Z. Liu, R. Laforest, J. Mhlanga, T. Fraum, M. Itani, F. Dehdashti, B. A. Siegel, A. K. Jha, “Observer study-based evaluation of a stochastic and physics-based method to generate oncological PET images” (link to proceedings and presentation) (arxiv) Z. Yu, […]

Four talks highlighted at SNMMI 2020!

Delighted to share that the four abstracts we had submitted to the SNMMI 2020 Annual Meeting were all accepted for talks and then also highlighted in the summary session of the PIDS council. The references are as below: 1. H.S. Moon, Z. Liu, M. Ponisio, R. Laforest, and A. K. Jha, A physics-guided and learning-based […]

Article on analysis of list-mode SPECT emission data accepted

Delighted to share that our article on analyzing the information content in SPECT emission data on the task of jointly estimating activity and attenuation distribution was accepted to Inverse Problems. Link to the article is here. Arxiv link here. This work shows that scatter-window photons in SPECT, that are typically discarded, when acquired in list-mode […]

Article on physics-guided AI-based segmentation method accepted

Excited to share that our article on developing a modular physics-guided deep-learning-based method for segmenting oncological PET images was recently accepted to the journal Physics in Medicine and Biology. Here is the link to the article, and here is the arxiv link. The article was also highlighted in the McKelvey Engineering newsletter. Source code for […]

Hae Sol defends his thesis

Congratulations to Hae Sol on defending his thesis on developing new image-analysis methods for DaTscan SPECT. His committee consisted of Drs. Joel Perlmutter, Richard Laforest, Dennis Barbour, and Abhinav Jha. Hae Sol is the first Masters student to defend from the lab. Exciting times!

Zekun awarded SNMMI Student grant

Zekun Li has been awarded the SNMMI Student Research Grant: Discovering Molecular Imaging. In this grant, Zekun will develop novel methods for SPECT reconstruction. Congratulations, Zekun! News article here:

CMI Lab at ISBI 2020

The CMI lab had a great presence at the IEEE International Symposium on Biomedical Imaging 2020. Lab members gave three talks on SPECT reconstruction and image analysis during a very well-attended virtual session on Computational Imaging for Nuclear Medicine. Asheq also received financial support to cover his registration costs at the meeting. Thanks ISBI organizing […]

Presentation at FDA workshop on role of AI in medical imaging

Dr. Jha from the CMI Lab gave a brief invited presentation at the  FDA public workshop on the Emerging Role of Artificial Intelligence in Medical Imaging. His presentation was titled “AI in Nuclear Medicine: Opportunities and Risks”. The talk was later highlighted in the day’s summary. The webcast for the presentation is available here (at […]

CMI Lab@WMIC 2019: Ziping & Hae Sol present.

The CMI lab had an excellent presence at the WMIC 2019 meeting. Ziping and Hae Sol presented their work on combining physics and machine learning for segmenting nuclear-medicine images. Dr. Jha gave an invited talk at the meeting. Overall, a great conference!

CMI Lab @WMIC 2019: Dr. Jha gives invited talk

At the WMIC 2019 meeting, Dr. Jha gave an invited talk on methods for quantitative medical imaging at the Imaging Data Science Forum. The focus of the talk was on new and improved methods for quantitative nuclear-medicine imaging, especially given the emergence of machine learning-based tools.

More accomplishments of summer interns

Qiye Tan and Jinxin Liu, summer interns at our lab, presented the outstanding work they have conducted during their stay here at a seminar yesterday. Best wishes to these two students as they go ahead in their careers.

Abstracts on ML-based segmentation techniques accepted @WMIC2019

Two abstracts based on our work on developing new maching-learning-based approaches to segment SPECT and PET images have been accepted to the World Molecular Imaging Congress (WMIC) 2019. These efforts were led by Ziping Liu and Hae Sol Moon, and were thanks in large part due to the collaboration with the physicians at WashU!

Honor from SNMMI

Dr. Jha has been selected as Ones to Watch: 30 early career researchers making a difference in 2019 by the Society of Nuclear Medicine and Molecular Imaging (SNMMI).

Ziping, PhD student, joins lab

Delighted to share that Ziping Liu, PhD student from the Department of Biomedical Engineering at Washington University, joined the lab. Ziping obtained his Bachelors from the Department of Biomedical Engineering at University of Rochester. His research interests are in Task-specific medical imaging, PET System Modeling, and Image Analysis. Welcome Ziping! 

Asheq, PhD student, joins lab

Excited to share that Md Ashequr Rahman, PhD student from the Image Science PhD program, has joined the Computational Medical Imaging Lab. Asheq obtained his Bachelors in Electrical and Electronic Engineering from Bangladesh University of Engineering and Technology, Bangladesh. Welcome Asheq!

Dr. Jha delivers keynote at Siteman Oncologic Imaging program

Dr. Abhinav Jha presented the keynote at the Siteman Cancer Center Oncologic Imaging Research Program Student Poster Session and Special lecture. His talk was titled  “Computational Medical Imaging for Precision Medicine” and discussed various computational-imaging, that included machine-learning and model-based methods, to improve precision medicine. Further, excited to share that Priyanshu Jain, a student member […]

NIBIB Trailblazer award on SPECT reconstruction

Our lab has been awarded the NIBIB Trailblazer award for a project on developing new methods for transmission-less attenuation compensation in SPECT imaging! In this project, we propose to design and validate novel methods for SPECT reconstruction that will not require an additional transmission scan (such as the CT scan) for attenuation compensation. For more […]


Our lab has been awarded the NVIDIA GPU grant! NVIDIA will provide us state-of-the-art Quadro P6000 card. Armed with 3,840 CUDA parallel processing cores, a peak single-precision performance of 12 TFLOPS, and a 24 GB GPU memory, this card will provide an excellent high-performance computing platform for several computational medical imaging solutions that we are […]

New paper on reconstruction method for FMT brain imaging

Our paper on developing a new reconstruction method for fluorescence molecular tomography in the context of transcranial imaging has been published in Biomedical Optics Express. The paper demonstrates that accurately modeling noise in FMT while accounting for sparsity leads to improved reconstruction performance!

Abstract on deep learning for PET image analysis accepted @ SNMMI

Our abstract titled “A deep-learning-based fully automated segmentation approach to delineate tumors in FDG-PET images of patients with lung cancer” was accepted for an oral presentation at the 2018 SNMMI Annual Meeting. In this work, we developed and demonstrated the efficacy of a deep learning-based method for tumor segmentation in PET images. Since these images are […]

Joining WashU as Assistant Professor of BME/Radiology

Dr. Abhinav Jha will be joining Washington University in St. Louis as an Assistant Professor of Biomedical Engineering in the School of Engineering and Applied Sciences and of the Mallinckrodt Institute of Radiology at the School of Medicine starting April 2018. His recruitment is part of the Imaging Sciences Initiative at WashU.

Three articles on developing observers for myocardial SPECT

Optimizing myocardial perfusion SPECT systems and methods for perfusion defect detection is highly significant. While mathematical observers exist for this purpose, they do not account for the variability in cardiac defect characteristics. In a series of articles, we investigate the causes for this issue and propose a novel mathematical observer that addresses this issue. X. […]

Two new journal articles on clinical quantitative PET

Metrics derived from quantitative PET, in particular intra-tumor heterogeneity, have sparked considerable interest as biomarkers for prognosis and predicting therapy response. In collaboration with radiologists at the Division of Nuclear Medicine, we have been investigating the role of intra-tumor heterogeneity in patients with different cancers. We recently published a couple of articles based on our research, […]

Three talks at SNMMI Annual Meeting 2017

We have had three talks accepted for presentation at the Society of Nuclear Medicine and Molecular Imaging (SNMMI) Annual Meeting 2017. A. K. Jha, C. Marcus, R. Wray. R. Subramaniam, and E. Frey, “A no-gold-standard framework to evaluate FDG-PET tumor-segmentation methods on the task of measuring prognostic biomarkers for lung-cancer treatment”, Therapy Center Young Investigator Award, […]

New article on ADC estimation in diffusion MRI

Our paper on developing a statistical technique to estimate a single ADC value from a lesion using diffusion MRI was recently published in the journal Magnetic Resonance in Medicine. Especially exciting that the proposed method was rigorous, computationally fast, easy-to-implement, convenient-to-use, and more accurate than state-of-the-art methods in clinical settings! This work was in collaboration with […]