Peer-Reviewed Publications

Jacob Haimson and Sarah S. Donaldson Professor and Professor, by courtesy, of Electrical Engineering


  • Deep Learning-Based Water-Fat Separation from Dual-Echo Chemical Shift-Encoded Imaging. Bioengineering (Basel, Switzerland) Wu, Y., Alley, M., Li, Z., Datta, K., Wen, Z., Sandino, C., Syed, A., Ren, H., Xing, L., Lustig, M., Pauly, J., Vasanawala, S. 2022; 9 (10)


    Conventional water-fat separation approaches suffer long computational times and are prone to water/fat swaps. To solve these problems, we propose a deep learning-based dual-echo water-fat separation method. With IRB approval, raw data from 68 pediatric clinically indicated dual echo scans were analyzed, corresponding to 19382 contrast-enhanced images. A densely connected hierarchical convolutional network was constructed, in which dual-echo images and corresponding echo times were used as input and water/fat images obtained using the projected power method were regarded as references. Models were trained and tested using knee images with 8-fold cross validation and validated on out-of-distribution data from the ankle, foot, and arm. Using the proposed method, the average computational time for a volumetric dataset with ~400 slices was reduced from 10 min to under one minute. High fidelity was achieved (correlation coefficient of 0.9969, l1 error of 0.0381, SSIM of 0.9740, pSNR of 58.6876) and water/fat swaps were mitigated. I is of particular interest that metal artifacts were substantially reduced, even when the training set contained no images with metallic implants. Using the models trained with only contrast-enhanced images, water/fat images were predicted from non-contrast-enhanced images with high fidelity. The proposed water-fat separation method has been demonstrated to be fast, robust, and has the added capability to compensate for metal artifacts.

    View details for DOI 10.3390/bioengineering9100579

    View details for PubMedID 36290546

  • An Efficient Framework for Video Documentation of Bladder Lesions for Cystoscopy: A Proof-of-Concept Study. Journal of medical systems Eminaga, O., Ge, T. J., Shkolyar, E., Laurie, M. A., Lee, T. J., Hockman, L., Jia, X., Xing, L., Liao, J. C. 2022; 46 (11): 73


    Processing full-length cystoscopy videos is challenging for documentation and research purposes. We therefore designed a surgeon-guided framework to extract short video clips with bladder lesions for more efficient content navigation and extraction. Screenshots of bladder lesions were captured during transurethral resection of bladder tumor, then manually labeled according to case identification, date, lesion location, imaging modality, and pathology. The framework used the screenshot to search for and extract a corresponding 10-seconds video clip. Each video clip included a one-second space holder with a QR barcode informing the video content. The success of the framework was measured by the secondary use of these short clips and the reduction of storage volume required for video materials. From 86 cases, the framework successfully generated 249 video clips from 230 screenshots, with 14 erroneous video clips from 8 screenshots excluded. The HIPPA-compliant barcodes provided information of video contents with a 100% data completeness. A web-based educational gallery was curated with various diagnostic categories and annotated frame sequences. Compared with the unedited videos, the informative short video clips reduced the storage volume by 99.5%. In conclusion, our framework expedites the generation of visual contents with surgeon's instruction for cystoscopy and potential incorporation of video data towards applications including clinical documentation, education, and research.

    View details for DOI 10.1007/s10916-022-01862-8

    View details for PubMedID 36190581

  • Radio-luminescent imaging for rapid, high resolution eye plaque loading verification. Medical physics Yan, H., De Jean, P., Grafil, E., Ashraf, R., Niedermayr, T., Astrahan, M., Mruthyunjaya, P., Beadle, B., Xing, L., Liu, W. 2022


    BACKGROUND: Eye plaque brachytherapy (EPB) is currently an optimal therapy for intraocular cancers. Due to the lack of an effective and practical technique to measure the seed radioactivity distribution, current quality assurance (QA) practice according to the AAPM TG129 only stipulates that the plaque assembly be visually inspected. Consequently, uniform seed activity is routinely adopted to avoid possible loading mistakes of differential seed loading. However, modulated dose delivery, which represents a general trend in radiotherapy to provide more personalized treatment for a given tumor and patient, requires differential activities in the loaded seeds.PURPOSE: In this study, a fast and low-cost radio-luminescent imaging and dose calculating system to verify the seed activity distribution for differential loading was developed.METHODS: A proof-of-concept system consisting of a thin scintillator sheet coupled to a camera/lens system was constructed. A seed-loaded plaque can be placed directly on the scintillator surface with the radioactive seeds facing the scintillator. The camera system collects the radioluminescent signal generated by the scintillator at its opposite side. The predicted dose distribution in the scintillator's sensitive layer was calculated using a Monte Carlo simulation with the planned plaque loading pattern of I-125 seeds. Quantitative comparisons of the distribution of relative measured signal intensity and that of the relative predicted dose in the sensitive layer were performed by gamma analysis, similar to IMRT QA.RESULTS: Data analyses showed high gamma (3%/0.3mm, global, 20% threshold) passing rates for correct seed loadings and low passing rates with distinguished high gamma value area for incorrect loadings, indicating that possible errors may be detected. The measurement and analysis only required a few extra minutes, significantly shorter than the time to assay the extra verification seeds the physicist already must perform as recommended by TG129.CONCLUSIONS: Radio-luminescent QA can be used to facilitate and assure the implementation of intensity modulated, customized plaque loading. This article is protected by copyright. All rights reserved.

    View details for DOI 10.1002/mp.16003

    View details for PubMedID 36183146

  • Mitigating the uncertainty in small field dosimetry by leveraging machine learning strategies. Physics in medicine and biology Zhao, W., Yang, Y., Xing, L., Chuang, C. F., Schüler, E. 2022


    Small field dosimetry is significantly different from the dosimetry of broad beams due to loss of electron side scatter equilibrium, source occlusion, and effects related to the choice of detector. However, use of small fields is increasing with the increase in indications for intensity-modulated radiation therapy (IMRT) and stereotactic body radiation therapy (SBRT), and thus the need for accurate dosimetry is ever more important. Here we propose to leverage machine learning (ML) strategies to reduce the uncertainties and increase the accuracy in determining small field output factors (OFs). Linac OFs from a Varian TrueBeam STx were calculated either by the treatment planning system (TPS) or measured with a W1 scintillator detector at various multi-leaf collimator (MLC) positions, jaw positions, and with and without contribution from leaf-end transmission. The fields were defined by the MLCs with the jaws at various positions. Field sizes between 5 and 100 mm were evaluated. Separate ML regression models were generated based on the TPS calculated or the measured datasets. Accurate predictions of small field OFs at different field sizes (FSs) were achieved independent of jaw and MLC position. A mean and maximum % relative error (RE) of 0.380.39% and 3.62%, respectively, for the best-performing models based on the measured datasets were found. The prediction accuracy was independent of contribution from leaf-end transmission. Several ML models for predicting small field OFs were generated, validated, and tested. Incorporating these models into the dose calculation workflow could greatly increase the accuracy and robustness of dose calculations for any radiotherapy delivery technique that relies heavily on small fields.

    View details for DOI 10.1088/1361-6560/ac7fd6

    View details for PubMedID 35803256

  • Shifting machine learning for healthcare from development to deployment and from models to data. Nature biomedical engineering Zhang, A., Xing, L., Zou, J., Wu, J. C. 2022


    In the past decade, the application of machine learning (ML) to healthcare has helped drive the automation of physician tasks as well as enhancements in clinical capabilities and access to care. This progress has emphasized that, from model development to model deployment, data play central roles. In this Review, we provide a data-centric view of the innovations and challenges that are defining ML for healthcare. We discuss deep generative models and federated learning as strategies to augment datasets for improved model performance, as well as the use of the more recent transformer models for handling larger datasets and enhancing the modelling of clinical text. We also discuss data-focused problems in the deployment of ML, emphasizing the need to efficiently deliver data to ML models for timely clinical predictions and to account for natural data shifts that can deteriorate model performance.

    View details for DOI 10.1038/s41551-022-00898-y

    View details for PubMedID 35788685