Publications by authors named "Ziyu Su"

Accurate recurrence risk stratification is crucial for optimizing treatment plans for breast cancer patients. Current prognostic tools like Oncotype DX (ODX) offer valuable genomic insights for HR+/HER2- patients but are limited by cost and accessibility, particularly in underserved populations. In this study, we present Deep-BCR-Auto, a deep learning-based computational pathology approach that predicts breast cancer recurrence risk from routine H&E-stained whole slide images (WSIs).

View Article and Find Full Text PDF
Article Synopsis
  • Lung cancer, particularly lung adenocarcinoma (LUAD), is the most common cause of cancer death in the U.S. and requires careful monitoring after surgical removal due to high recurrence risk.
  • Traditional methods of assessing tumor grading are labor-intensive and may vary between observers, leading to inconsistencies in treatment planning.
  • This study introduces a deep learning model with a dual-attention architecture that effectively predicts the 5-year recurrence risk of LUAD post-surgery, outperforming existing methods and enhancing decision-making in patient treatment.
View Article and Find Full Text PDF

Although multiple instance learning (MIL) methods are widely used for automatic tumor detection on whole slide images (WSI), they suffer from the extreme class imbalance WSIs containing small tumors where the tumor may include only a few isolated cells. For early detection, it is important that MIL algorithms can identify small tumors. Existing studies have attempted to address this issue using attention-based architectures and instance selection-based methodologies but have not produced significant improvements.

View Article and Find Full Text PDF

Colorectal cancer (CRC) is the third most common cancer in the United States. Tumor Budding (TB) detection and quantification are crucial yet labor-intensive steps in determining the CRC stage through the analysis of histopathology images. To help with this process, we adapt the Segment Anything Model (SAM) on the CRC histopathology images to segment TBs using SAM-Adapter.

View Article and Find Full Text PDF

Current deep learning methods in histopathology are limited by the small amount of available data and time consumption in labeling the data. Colorectal cancer (CRC) tumor budding quantification performed using H&E-stained slides is crucial for cancer staging and prognosis but is subject to labor-intensive annotation and human bias. Thus, acquiring a large-scale, fully annotated dataset for training a tumor budding (TB) segmentation/detection system is difficult.

View Article and Find Full Text PDF

Tumor budding refers to a cluster of one to four tumor cells located at the tumor-invasive front. While tumor budding is a prognostic factor for colorectal cancer, counting and grading tumor budding are time consuming and not highly reproducible. There could be high inter- and intra-reader disagreement on H&E evaluation.

View Article and Find Full Text PDF

Conventional spectroscopies are not sufficiently selective to comprehensively understand the behaviour of trapped carriers in perovskite solar cells, particularly under their working conditions. Here we use infrared optical activation spectroscopy (i.e.

View Article and Find Full Text PDF
Article Synopsis
  • Current multiple instance learning (MIL) models struggle with whole slide images (WSIs) that contain tiny tumor lesions due to the small tumor-to-normal area ratio, which hinders accurate classification.
  • The proposed salient instance inference MIL (SiiMIL) model improves WSI classification by introducing a representation learning method that highlights key normal instances, allowing for better selection of tumor instances and higher tumor-to-normal ratios in data bags.
  • SiiMIL achieves notable performance with a 0.9225 AUC and 0.7551 recall on the Camelyon16 dataset, outperforming existing models and providing more interpretable tumor-sensitive attention heatmaps for pathologists.
View Article and Find Full Text PDF

Artificial Intelligence (AI)-enhanced histopathology presents unprecedented opportunities to benefit oncology through interpretable methods that require only one overall label per hematoxylin and eosin (H&E) slide with no tissue-level annotations. We present a structured review of these methods organized by their degree of verifiability and by commonly recurring application areas in oncological characterization. First, we discuss morphological markers (tumor presence/absence, metastases, subtypes, grades) in which AI-identified regions of interest (ROIs) within whole slide images (WSIs) verifiably overlap with pathologist-identified ROIs.

View Article and Find Full Text PDF

Obstructive sleep apnea (OSA) is a prevalent disease affecting 10 to 15% of Americans and nearly one billion people worldwide. It leads to multiple symptoms including daytime sleepiness; snoring, choking, or gasping during sleep; fatigue; headaches; non-restorative sleep; and insomnia due to frequent arousals. Although polysomnography (PSG) is the gold standard for OSA diagnosis, it is expensive, not universally available, and time-consuming, so many patients go undiagnosed due to lack of access to the test.

View Article and Find Full Text PDF

The early diagnosis of lymph node metastasis in breast cancer is essential for enhancing treatment outcomes and overall prognosis. Unfortunately, pathologists often fail to identify small or subtle metastatic deposits, leading them to rely on cytokeratin stains for improved detection, although this approach is not without its flaws. To address the need for early detection, multiple-instance learning (MIL) has emerged as the preferred deep learning method for automatic tumor detection on whole slide images (WSIs).

View Article and Find Full Text PDF

Breast cancer is the most common malignancy in women, with over 40,000 deaths annually in the United States alone. Clinicians often rely on the breast cancer recurrence score, Oncotype DX (ODX), for risk stratification of breast cancer patients, by using ODX as a guide for personalized therapy. However, ODX and similar gene assays are expensive, time-consuming, and tissue destructive.

View Article and Find Full Text PDF

Deep learning consistently demonstrates high performance in classifying and segmenting medical images like CT, PET, and MRI. However, compared to these kinds of images, whole slide images (WSIs) of stained tissue sections are huge and thus much less efficient to process, especially for deep learning algorithms. To overcome these challenges, we present attention2majority, a weak multiple instance learning model to automatically and efficiently process WSIs for classification.

View Article and Find Full Text PDF
Article Synopsis
  • The CADA challenge aimed to improve algorithms for detecting and analyzing cerebral aneurysms in 3D rotational angiography images by providing training on 109 anonymized datasets and testing on 22 additional ones.
  • Participants from 22 countries created detection solutions primarily using U-Net, achieving a high F2 score of 0.92, which is comparable to expert performance, though smaller aneurysms were sometimes missed.
  • The challenge also assessed rupture risk estimation, with the best methods combining various parameters to achieve an F2 score of 0.70, closely matching the 0.71 score when using expert-defined structures.
View Article and Find Full Text PDF