Medical-image Analysis and Statistical Interpretation (MASI) Lab

Validation and Optimization of Multi-Organ Segmentation on Clinical Imaging Archives

Feb. 12, 2020—Olivia Tang, Yuchen Xu, Yucheng Tang, Ho Hin Lee, Yunqiang Chen, Dashan Gao, Shizhong Han, Riqiang Gao, Michael R. Savona, Richard G. Abramson, Yuankai Huo, Bennett A. Landman, “Validation and Optimization of Multi-Organ Segmentation on Clinical Imaging Archives”, SPIE IP:MI 2020. Houston, TX. https://arxiv.org/abs/2002.04102 Abstract Segmentation of abdominal computed tomography (CT) provides spatial context, morphological...

Examples of segmentations generated by the baseline algorithm. A. An inlier, where the algorithm correctly predicted larger organs, like the liver (on the left side of image in purple) and spleen (on the right side of image in red- pink) and suggested mostly accurate areas of smaller organs. B. An outlier (global failure), where the liver (purple) and spleen (red-pink) were largely correct but major inconsistencies are visible with other organs.

Outlier Guided Optimization of Abdominal Segmentation

Feb. 12, 2020—Yuchen Xu*, Olivia Tang*, Yucheng Tang, Ho Hin Lee, Yunqiang Chen, Dashan Gao, Shizhong Han, Riqiang Gao, Michael R. Savona, Richard G. Abramson, Yuankai Huo, Bennett A. Landman, “Outlier Guided Optimization of Abdomen Segmentation”, SPIE IP:MI 2020. Houston, TX https://arxiv.org/abs/2002.04098 Abstract Abdominal multi-organ segmentation of computed tomography (CT) images has been the subject of extensive...

Different classes of methods have been used to infer tissue microstructure from single shell and multi shell DW-MRI data. The gap addressed herein is in data-driven machine learning models for multi-shell DW-MRI data.

Enabling Multi-shell b-Value Generalizability of Data-Driven Diffusion Models with Deep SHORE

Jan. 17, 2020—Nath V, Lyu I, Schilling KG, Parvathaneni P, Hansen CB, Huo Y, Janve VA, Gao Y, Stepniewska I, Anderson AW, Landman BA. Enabling Multi-shell b-Value Generalizability of Data-Driven Diffusion Models with Deep SHORE. In International Conference on Medical Image Computing and Computer-Assisted Intervention 2019 Oct 13 (pp. 573-581). Springer, Cham. Full text: https://arxiv.org/ftp/arxiv/papers/1907/1907.06319.pdf Abstract Intra-voxel...

The base regression neural network is depicted in the center which describes the parameters of the fullyconnected dense layers with respective activation functions. The plot at left: Depicts three inputs where the center input
comes from corresponding DW-MRI with histology. The other two are pairwise inputs from corresponding voxels of
scanner 1.5T and 3T. The plot at right: Depicts the loss function which uses the hypothesis that the outcome/prediction
should be same irrespective of the scanner gradient strength.

Harmonizing 1.5 T/3T diffusion weighted MRI through development of deep learning stabilized microarchitecture estimators

Jan. 17, 2020—Nath V, Remedios S, Parvathaneni P, Hansen CB, Bayrak RG, Bermudez C, Blaber JA, Schilling KG, Janve VA, Gao Y, Huo Y. Harmonizing 1.5 T/3T diffusion weighted MRI through development of deep learning stabilized microarchitecture estimators. In Medical Imaging 2019: Image Processing 2019 Mar 15 (Vol. 10949, p. 109490O). International Society for Optics and Photonics....

Violin plots of intrasession submissions across both the scanners per tract. (A) Dice similarity coefficients, (B) intraclass correlation coefficients. The top row depicts the median of the top five intrasession submissions. The tracts are in the following order (L/R): a) Uncinate, b) Fornix, c) Fminor & Fmajor, d) Cingulum, e) Corticospinal tract, f) Inferior longitudinal fasciculus, g) Superior longitudinal fasciculus, h) Inferior fronto‐occipital tract.

Tractography reproducibility challenge with empirical data (TRAceD): The 2017 ISMRM diffusion study group challenge

Jan. 17, 2020—Nath V, Schilling KG, Parvathaneni P, Huo Y, Blaber JA, Hainline AE, Barakovic M, Romascano D, Rafael‐Patino J, Frigo M, Girard G. Tractography reproducibility challenge with empirical data (traced): The 2017 ISMRM diffusion study group challenge. Journal of Magnetic Resonance Imaging. 2020 Jan;51(1):234-49. Full text: https://www.ncbi.nlm.nih.gov/pubmed/31179595 Abstract BACKGROUND: Fiber tracking with diffusion-weighted MRI has become an...

A) Middle axial slice of a human brain acquired using DW-MRI. B, C) CSD FOD's of the same subject on two different scanners showing inconsistency which cannot be validated without ground truth. D) Coronal slice of a Squirrel Monkey. E, F, G, H) Right: CSD FOD's at b-value of 3000 s/mm2 Left: Ground truth reconstruction using aggregated histological structural tensors depicting a loss in precision.

Deep learning reveals untapped information for local white-matter fiber reconstruction in diffusion-weighted MRI

Jan. 17, 2020—Nath V, Schilling KG, Parvathaneni P, Hansen CB, Hainline AE, Huo Y, Blaber JA, Lyu I, Janve V, Gao Y, Stepniewska I, Anderson AW, Landman BA. Deep learning reveals untapped information for local white-matter fiber reconstruction in diffusion-weighted MRI. Magnetic resonance imaging. 2019 Oct 1;62:220-7. Abstract PURPOSE: Diffusion-weighted magnetic resonance imaging (DW-MRI) is of critical importance...

Figure 2

Illustration of multiple instance learning with gradient accumulation. The first CT volume is organized into a bag of its 2D slices. Then, the model performs inference on all slices in the bag, and class probabilities are calculated. The gradient is calculated only for the instance corresponding to the most probable positive class. This gradient is saved and the gradient calculation is repeated for the next bag until the batch is done, and the accumulated gradient is averaged for the batch size and applied to the model.

Extracting 2D weak labels from volume labels using multiple instance learning in CT hemorrhage detection

Jan. 2, 2020—Remedios, S. W., Wu, Z., Bermudez, C., Kerley, C. I., Roy, S., Patel, M. B., Butman, J. A., Landman, B. A., Pham, D. L. (2019). Extracting 2D weak labels from volume labels using multiple instance learning in CT hemorrhage detection. arXiv preprint arXiv:1911.05650. Full Text: Arxiv Link Abstract Multiple instance learning (MIL) is a supervised learning methodology...

Distributed Deep Learning Across Multisite Datasets for Generalized CT Hemorrhage Segmentation

Jan. 2, 2020—Remedios, S. W., Roy, S., Bermudez, C., Patel, M. B., Butman, J. A., Landman, B. A., & Pham, D. L. (2019). Distributed Deep Learning Across Multi‐site Datasets for Generalized CT Hemorrhage Segmentation. Medical physics. Full Text: Pubmed Link Abstract Purpose: As deep neural networks achieve more success in the wide field of computer vision, greater emphasis is...

Distributed deep learning for robust multi-site segmentation of CT imaging after traumatic brain injury

Jan. 2, 2020—Remedios, Samuel, et al. “Distributed deep learning for robust multi-site segmentation of CT imaging after traumatic brain injury.” Medical Imaging 2019: Image Processing. Vol. 10949. International Society for Optics and Photonics, 2019. Full text: PubMed Link Abstract Machine learning models are becoming commonplace in the domain of medical imaging, and with these methods comes an ever-increasing need...

Semi-Supervised Multi-Organ Segmentation through Quality Assurance Supervision

Dec. 19, 2019—Ho Hin Lee, Yucheng Tang, Olivia Tang, Yuchen Xu, Yunqiang Chen, Dashan Gao, Shizhong Han, Riqiang Gao, Michael R. Savona, Richard G. Abramson, Yuankai Huo, Bennett A. Landman, “Semi-Supervised Multi-Organ Segmentation through Quality Assurance Supervision”, SPIE MI:IP 2020. Houston, TX. Link: https://arxiv.org/abs/1911.05113 Abstract Human in-the-loop quality assurance (QA) is typically performed after medical image segmentation to ensure that...

« Previous Page