1 code implementation • 15 Apr 2024 • Hanxue Gu, Haoyu Dong, Jichen Yang, Maciej A. Mazurowski
Automated segmentation is a fundamental medical image analysis task, which enjoys significant advances due to the advent of deep learning.
no code implementations • 10 Apr 2024 • Nicholas Konz, YuWen Chen, Hanxue Gu, Haoyu Dong, Maciej A. Mazurowski
Modern medical image translation methods use generative models for tasks such as the conversion of CT images to MRI.
no code implementations • 18 Mar 2024 • Keyu Li, Hanxue Gu, Roy Colglazier, Robert Lark, Elizabeth Hubbard, Robert French, Denise Smith, Jikai Zhang, Erin McCrum, Anthony Catanzano, Joseph Cao, Leah Waldman, Maciej A. Mazurowski, Benjamin Alman
To address these challenges and the lack of interpretability found in certain existing automated methods, we have created fully automated software that not only precisely measures the Cobb angle but also provides clear visualizations of these measurements.
no code implementations • 16 Mar 2024 • YuWen Chen, Nicholas Konz, Hanxue Gu, Haoyu Dong, Yaqian Chen, Lin Li, Jisoo Lee, Maciej A. Mazurowski
We evaluate our method by training a segmentation model on images translated from CT to MRI with their original CT masks and testing its performance on real MRIs.
1 code implementation • 14 Feb 2024 • Haoyu Dong, Nicholas Konz, Hanxue Gu, Maciej A. Mazurowski
Here, we approach such a task, of adapting a medical image segmentation model with only a single unlabeled test image.
1 code implementation • 7 Feb 2024 • Nicholas Konz, YuWen Chen, Haoyu Dong, Maciej A. Mazurowski
Diffusion models have enabled remarkably high-quality medical image generation, yet it is challenging to enforce anatomical constraints in generated images.
1 code implementation • 23 Jan 2024 • Hanxue Gu, Roy Colglazier, Haoyu Dong, Jikai Zhang, Yaqian Chen, Zafer Yildiz, YuWen Chen, Lin Li, Jichen Yang, Jay Willhite, Alex M. Meyer, Brian Guo, Yashvi Atul Shah, Emily Luo, Shipra Rajput, Sally Kuehn, Clark Bulleit, Kevin A. Wu, Jisoo Lee, Brandon Ramirez, Darui Lu, Jay M. Levin, Maciej A. Mazurowski
In our study, we propose a versatile, publicly available deep-learning model for bone segmentation in MRI across multiple standard MRI locations.
1 code implementation • 16 Jan 2024 • Nicholas Konz, Maciej A. Mazurowski
We address this gap in knowledge by establishing and empirically validating a generalization scaling law with respect to $d_{data}$, and propose that the substantial scaling discrepancy between the two considered domains may be at least partially attributed to the higher intrinsic ``label sharpness'' ($K_\mathcal{F}$) of medical imaging datasets, a metric which we propose.
no code implementations • 17 Dec 2023 • Yixin Zhang, Shen Zhao, Hanxue Gu, Maciej A. Mazurowski
In situations where unlimited annotation time was available, precise annotations still lead to the highest segmentation model performance.
no code implementations • 5 Oct 2023 • Jee Seok Yoon, Kwanseok Oh, Yooseung Shin, Maciej A. Mazurowski, Heung-Il Suk
Medical image analysis (MedIA) has become an essential tool in medicine and healthcare, aiding in disease diagnosis, prognosis, and treatment planning, and recent successes in deep learning (DL) have made significant contributions to its advances.
no code implementations • 28 Jun 2023 • Hanxue Gu, Haoyu Dong, Nicholas Konz, Maciej A. Mazurowski
We experimentally study the effects of different aspects of F-B imbalance (object size, number of objects, dataset size, object type) on detection performance.
no code implementations • 11 May 2023 • Yixin Zhang, Maciej A. Mazurowski
Shape learning, or the ability to leverage shape information, could be a desirable property of convolutional neural networks (CNNs) when target objects have specific shapes.
1 code implementation • 4 May 2023 • Nicholas Konz, Haoyu Dong, Maciej A. Mazurowski
Given the scarcity of abnormal images and the abundance of normal images for this problem, an anomaly detection/localization approach could be well-suited.
2 code implementations • 20 Apr 2023 • Maciej A. Mazurowski, Haoyu Dong, Hanxue Gu, Jichen Yang, Nicholas Konz, Yixin Zhang
We conclude that SAM shows impressive zero-shot segmentation performance for certain medical imaging datasets, but moderate to poor performance for others.
1 code implementation • 8 Mar 2023 • Nicholas Konz, Maciej A. Mazurowski
The image acquisition parameters (IAPs) used to create MRI scans are central to defining the appearance of the images.
1 code implementation • 5 Jan 2023 • Shixing Cao, Nicholas Konz, James Duncan, Maciej A. Mazurowski
In this work we introduce a novel medical image style transfer method, StyleMapper, that can transfer medical scans to an unseen style with access to limited training data.
no code implementations • 27 Jul 2022 • Jingxi Weng, Benjamin Wildman-Tobriner, Mateusz Buda, Jichen Yang, Lisa M. Ho, Brian C. Allen, Wendy L. Ehieli, Chad M. Miller, Jikai Zhang, Maciej A. Mazurowski
Objectives: The purpose is to apply a previously validated deep learning algorithm to a new thyroid nodule ultrasound image dataset and compare its performances with radiologists.
no code implementations • 25 Jul 2022 • Albert Swiecicki, Nianyi Li, Jonathan O'Donnell, Nicholas Said, Jichen Yang, Richard C. Mather, William A. Jiranek, Maciej A. Mazurowski
A novel deep learning-based method was utilized for assessment of knee OA in two steps: (1) localization of knee joints in the images, (2) classification according to the KL grading system.
1 code implementation • 6 Jul 2022 • Nicholas Konz, Hanxue Gu, Haoyu Dong, Maciej A. Mazurowski
These results give a more principled underpinning for the intuition that radiological images can be more challenging to apply deep learning to than natural image datasets common to machine learning research.
1 code implementation • 16 Mar 2022 • Hanxue Gu, Keyu Li, Roy J. Colglazier, Jichen Yang, Michael Lebhar, Jonathan O'Donnell, William A. Jiranek, Richard C. Mather, Rob J. French, Nicholas Said, Jikai Zhang, Christine Park, Maciej A. Mazurowski
We propose a novel deep learning-based five-step algorithm to automatically grade KOA from posterior-anterior (PA) views of radiographs: (1) image preprocessing (2) localization of knees joints in the image using the YOLO v3-Tiny model, (3) initial assessment of the severity of osteoarthritis using a convolutional neural network-based classifier, (4) segmentation of the joints and calculation of the joint space narrowing (JSN), and (5), a combination of the JSN and the initial assessment to determine a final Kellgren-Lawrence (KL) score.
no code implementations • 7 Mar 2022 • Fakrul Islam Tushar, Ehsan Abadi, Saman Sotoudeh-Paima, Rafael B. Fricks, Maciej A. Mazurowski, W. Paul Segars, Ehsan Samei, Joseph Y. Lo
However, performance dropped to an AUC of 0. 65 and 0. 69 when evaluated on clinical and our simulated CVIT-COVID dataset.
no code implementations • 3 Mar 2022 • Fakrul Islam Tushar, Husam Nujaim, Wanyi Fu, Ehsan Abadi, Maciej A. Mazurowski, Ehsan Samei, William P. Segars, Joseph Y. Lo
This demonstrates that quality data is the key to improving the model's performance.
no code implementations • 22 Nov 2021 • Yifan Zhang, Haoyu Dong, Nicholas Konz, Hanxue Gu, Maciej A. Mazurowski
Specifically, we propose a novel modification of visual transformer (ViT) on image feature patches to connect the feature patches of a tumor with healthy backgrounds of breast images and form a more robust backbone for tumor detection.
1 code implementation • 5 Feb 2021 • Vincent M. D'Anniballe, Fakrul Islam Tushar, Khrystyna Faryna, Songyue Han, Maciej A. Mazurowski, Geoffrey D. Rubin, Joseph Y. Lo
Pre-trained models outperformed random initialization across all diseases.
1 code implementation • 13 Nov 2020 • Mateusz Buda, Ashirbani Saha, Ruth Walsh, Sujata Ghate, Nianyi Li, Albert Święcicki, Joseph Y. Lo, Maciej A. Mazurowski
While breast cancer screening has been one of the most studied medical imaging applications of artificial intelligence, the development and evaluation of the algorithms are hindered due to the lack of well-annotated large-scale publicly available datasets.
1 code implementation • 31 Oct 2020 • Anindo Saha, Fakrul I. Tushar, Khrystyna Faryna, Vincent M. D'Anniballe, Rui Hou, Maciej A. Mazurowski, Geoffrey D. Rubin, Joseph Y. Lo
Second, segmentation and classification models are connected with two different feature aggregation strategies to enhance the classification performance.
1 code implementation • 3 Aug 2020 • Fakrul Islam Tushar, Vincent M. D'Anniballe, Rui Hou, Maciej A. Mazurowski, Wanyi Fu, Ehsan Samei, Geoffrey D. Rubin, Joseph Y. Lo
Purpose: To design multi-disease classifiers for body CT scans for three different organ systems using automatically extracted labels from radiology text reports. Materials & Methods: This retrospective study included a total of 12, 092 patients (mean age 57 +- 18; 6, 172 women) for model development and testing (from 2012-2017).
1 code implementation • 12 Feb 2020 • Rachel Lea Draelos, David Dov, Maciej A. Mazurowski, Joseph Y. Lo, Ricardo Henao, Geoffrey D. Rubin, Lawrence Carin
This model reached a classification performance of AUROC greater than 0. 90 for 18 abnormalities, with an average AUROC of 0. 773 for all 83 abnormalities, demonstrating the feasibility of learning from unfiltered whole volume CT data.
no code implementations • 16 Dec 2019 • Gourav Modanwal, Adithya Vellal, Maciej A. Mazurowski
Second, we propose a modified discriminator architecture which utilizes a smaller field-of-view to ensure the preservation of finer details in the breast tissue.
4 code implementations • 9 Jun 2019 • Mateusz Buda, Ashirbani Saha, Maciej A. Mazurowski
Based on automatic deep learning segmentations, we extracted three features which quantify two-dimensional and three-dimensional characteristics of the tumors.
Ranked #2 on Brain Segmentation on Brain MRI segmentation
no code implementations • 5 Jul 2018 • Jun Zhang, Ashirbani Saha, Brian J. Soher, Maciej A. Mazurowski
Then, based on the segmentation results, a subject-specific piecewise linear mapping function was applied between the anchor points to normalize the same type of tissue in different patients into the same intensity ranges.
no code implementations • 10 Feb 2018 • Maciej A. Mazurowski, Mateusz Buda, Ashirbani Saha, Mustafa R. Bashir
In this article, we review the clinical reality of radiology and discuss the opportunities for application of deep learning algorithms.
no code implementations • 29 Nov 2017 • Zhe Zhu, Ehab AlBadawy, Ashirbani Saha, Jun Zhang, Michael R. Harowicz, Maciej A. Mazurowski
Results: The best AUC performance for distinguishing molecular subtypes was 0. 65 (95% CI:[0. 57, 0. 71]) and was achieved by the off-the-shelf deep features approach.
no code implementations • 28 Nov 2017 • Zhe Zhu, Michael Harowicz, Jun Zhang, Ashirbani Saha, Lars J. Grimm, E. Shelley Hwang, Maciej A. Mazurowski
In the first approach, we adopted the transfer learning strategy, in which a network pre-trained on a large dataset of natural images is fine-tuned with our DCIS images.
3 code implementations • 15 Oct 2017 • Mateusz Buda, Atsuto Maki, Maciej A. Mazurowski
In our study, we use three benchmark datasets of increasing complexity, MNIST, CIFAR-10 and ImageNet, to investigate the effects of imbalance on classification and perform an extensive comparison of several methods to address the issue: oversampling, undersampling, two-phase training, and thresholding that compensates for prior class probabilities.