Reducing annotation burden in MR: A novel MR-contrast guided contrastive learning approach for image segmentation.

BACKGROUND: Contrastive learning, a successful form of representational learning, has shown promising results in pretraining deep learning (DL) models for downstream tasks. When working with limited annotation data, as in medical image segmentation tasks, learning domain-specific local representations can further improve the performance of DL models. PURPOSE: In this work, we extend the contrastive learning framework to utilize domain-specific contrast information from unlabeled Magnetic Resonance (MR) images to improve the performance of downstream MR image segmentation tasks in the presence of limited labeled data. METHODS: The contrast in MR images is controlled by underlying tissue properties (e.g., T1 or T2) and image acquisition parameters. We hypothesize that learning to discriminate local representations based on underlying tissue properties should improve subsequent segmentation tasks on MR images. We propose a novel constrained contrastive learning (CCL) strategy that uses tissue-specific information via a constraint map to define positive and negative local neighborhoods for contrastive learning, embedding this information in the representational space during pretraining. For a given MR contrast image, the proposed strategy uses local signal characteristics (constraint map) across a set of related multi-contrast MR images as a surrogate for underlying tissue information. We demonstrate the utility of the approach for downstream: (1) multi-organ segmentation tasks in T2-weighted images where a DL model learns T2 information with constraint maps from a set of 2D multi-echo T2-weighted images (n = 101) and (2) tumor segmentation tasks in multi-parametric images from the public brain tumor segmentation (BraTS) (n = 80) dataset where DL models learn T1 and T2 information from multi-parametric BraTS images. Performance is evaluated on downstream multi-label segmentation tasks with limited data in (1) T2-weighted images of the abdomen from an in-house Radial-T2 (Train/Test = 30/20), (2) public Cartesian-T2 (Train/Test = 6/12) dataset, and (3) multi-parametric MR images from the public brain tumor segmentation dataset (BraTS) (Train/Test = 40/50). The performance of the proposed CCL strategy is compared to state-of-the-art self-supervised contrastive learning techniques. In each task, a model is also trained using all available labeled data for supervised baseline performance. RESULTS: The proposed CCL strategy consistently yielded improved Dice scores, Precision, and Recall metrics, and reduced HD95 values across all segmentation tasks. We also observed performance comparable to the baseline with reduced annotation effort. The t-SNE visualization of features for T2-weighted images demonstrates its ability to embed T2 information in the representational space. On the BraTS dataset, we also observed that using an appropriate multi-contrast space to learn T1+T2, T1, or T2 information during pretraining further improved the performance of tumor segmentation tasks. CONCLUSIONS: Learning to embed tissue-specific information that controls MR image contrast with the proposed constrained contrastive learning improved the performance of DL models on subsequent segmentation tasks compared to conventional self-supervised contrastive learning techniques. The use of such domain-specific local representations could help understand, improve performance, and mitigate the scarcity of labeled data in MR image segmentation tasks.

Long-Term Follow-up of High-Dose Chemotherapy with Autologous Stem Cell Transplantation in Children and Young Adults with Metastatic or Relapsed Ewing Sarcoma: A Single-Institution Experience.

Forty-seven patients with metastatic disease at diagnosis or recurrent Ewing sarcoma (EWS) received high-dose chemotherapy (HDC) followed by tandem (n = 20, from February 13, 1997, to October 24, 2002) or single (n = 27, from October 1, 2004, to September 5, 2018) autologous hematopoietic stem cell transplantation (ASCT). To our knowledge, this is the largest single-institution study with sustained long-term follow-up exceeding 10 years. All patients who underwent single ASCT received a novel conditioning regimen with busulfan, melphalan, and topotecan. The overall survival (OS) and disease-free survival (DFS) were 46% and 37% at 10 years and 42% and 37% at 15 years, respectively. Disease status at transplant and the time to disease relapse prior to ASCT were identified as important prognostic factors in OS, DFS, and risk of relapse. At 10 years, patients who underwent transplantation in first complete response (1CR) had an excellent outcome (OS 78%), patients in 1CR/second complete response (2CR)/first partial response (1PR) had an OS of 66%, and patients at third or more complete response, second or more partial response, or advanced disease had an OS of 26%. Ten-year OS for patients without a history of relapse, with late relapse (≥2 years from diagnosis), or with early relapse (<2 years from diagnosis) was 75%, 50%, and 18%, respectively. Selected patients in 1CR, 2CR, 1PR, and with late relapse had excellent, sustained 10- and 15-year OS and DFS.