News

AI-driven medical imaging has emerged as a transformative force in modern healthcare, empowering clinicians to deliver more accurate, efficient, and personalized diagnostic and therapeutic strategies. By automatically analyzing CT, MRI, ultrasound, and other imaging modalities, AI systems can identify subtle patterns, precisely segment anatomical structures, and support clinical decision-making with enhanced accuracy and consistency. These advancements not only improve diagnostic performance but also streamline clinical workflows and ultimately elevate the overall quality of patient care.

Among the many tasks in medical image analysis, kidney and kidney tumor segmentation are pivotal for the management of renal diseases, particularly renal cell carcinoma. Precise delineation of kidneys and tumors is essential for quantitative tumor assessment, treatment planning, surgical navigation, postoperative monitoring, and radiomics research. Accurate segmentation enables clinicians to reliably estimate tumor burden, evaluate tumor–organ spatial relationships, and facilitate nephron-sparing surgical strategies, all of which directly influence patient outcomes. Given that manual segmentation is labor-intensive and prone to inter- and intra-observer variability, the development of automated, robust, and reliable segmentation methods has become increasingly critical for both routine clinical practice and large-scale research.

To address these limitations, we propose 3D-Cube Multi-Stage Green U-shaped Learning (GUSL), a novel multi-stage feed-forward machine learning framework for 3D medical image segmentation without backpropagation. GUSL is designed to be computationally efficient, interpretable, and environmentally sustainable, while maintaining competitive segmentation performance.

The proposed framework adopts a cascaded multi-stage segmentation strategy tailored to different anatomical tasks. As illustrated in Figure 1, distinct stages are designed for coarse-to-fine segmentation. First, the original CT volume is downsampled to a lower resolution, enabling efficient coarse localization of the kidney with reduced computational complexity. This low-resolution stage provides an approximate spatial position of the kidney, as highlighted in the red box. [...]

Histopathologic analysis is a key confirmatory step in the cancer diagnosis pipeline, where pathologists analyze a tissue section of interest for abnormalities and the extent of disease progression. The digitization of these tissue slides has enabled the use of AI for Whole Slide Image (WSI) Analysis, primarily to simplify pathologists’ laborious tasks and improve diagnostic accuracy. Due to the large size of these images, they’re split into smaller patches, and after analysis of individual patches, the results are aggregated to get the result for the WSI. Such a paradigm is called Multiple Instance Learning (MIL).

Architectural patterns surrounding the tumor regions are key indicators of angiogenesis and help in prognosis prediction. Architectural patterns are classified into 9 types and are grouped into three categories based on the underlying vasculature. While some patterns are often seen, others are rare and common only in higher-grade tumors. This causes a data imbalance that may affect training. To overcome this challenge, we propose an ensemble classifier for architectural pattern classification.

To capture local details and global context, we employ a multi-resolution feature encoder. At each resolution, the Saab transform is applied to obtain joint spatial-spectral representations. Representation learning is followed by a pooling operation to obtain compact representations. The pooled features from each resolution are concatenated to obtain a single feature vector, which is used to select the most discriminant features for the target classification task. An XGBoost binary classifier trained on these selected features predicts a confidence score for each architectural pattern. The confidence scores from multiple one-vs-one classifiers are aggregated to predict the architectural pattern in each patch.

One of the key components of Green Image Coding (GIC) is multi-grid control, which enables efficient and scalable bit allocation across the framework’s hierarchical layers. Unlike traditional hybrid codecs designed for single-layer encoding, GIC decomposes images into multiple hierarchical layers via resampling, referred to as a multi-grid representation. This decomposition effectively redistributes energy and reduces intra-layer content diversity, but it also creates a complex high-dimensional optimization challenge when attempting to allocate bits optimally across these various layers.

To make this problem tractable, we establish a theoretical foundation by defining the relationship between local and global rate-distortion (RD) models. We demonstrate that the global RD model can be derived from the local RD model of an individual layer by applying specific offsets to both rate and distortion. Notably, the distortion offset is a constant value determined by up-sampling processes and is unrelated to the compression process itself. This theoretical breakthrough reduces an intractable high-dimensional problem into a set of manageable sequential decisions.

Based on these findings, GIC implements a practical slope-matching-based rate control strategy. This strategy allocates bits across multiple grids by matching the slopes of consecutive RD curves. A primary advantage of this design is its modularity; the rate control module only requires information from two consecutive layers to function. This allows the module to be easily duplicated for any number of layers in the encoder, effectively decomposing the global rate-distortion optimization into a sequence of local optimizations to ensure a scalable balance between bit rate and image distortion.

Congratulations to Jiaxin Yang for passing his Qualifying Exam! His thesis proposal is titled “Green U-Shaped Learning for Medical Image Analysis: Methodologies and Applications.” His Qualifying Exam Committee members include Jay Kuo (Chair), Justin Haldar, Peter Beerel, Vasileios Magoulianitis, and Michael Khoo (Outside Member).

Artificial intelligence (AI) has rapidly transformed medical imaging by enabling more accurate, efficient, and personalized diagnosis and treatment. In particular, AI-driven medical image segmentation plays a critical role in clinical decision-making for diseases such as prostate cancer and renal cell carcinoma. However, most existing deep learning–based segmentation models rely heavily on backpropagation and large-scale computation, making them energy-intensive, difficult to interpret, and challenging to deploy in resource-constrained clinical settings.

To address these limitations, this work introduces Green U-shaped Learning (GUSL), a novel feed forward machine learning framework for 3D medical image segmentation without backpropagation. GUSL is designed to be efficient, interpretable, and environmentally sustainable, while maintaining competitive segmentation performance.

The proposed framework adopts cascaded multi-stage segmentation strategies tailored to different anatomical tasks. For prostate segmentation, a two-stage coarse-to-fine approach first localizes the prostate gland and then refines its boundaries, effectively mitigating severe class imbalance and anatomical variability, as shown in Figure 1. For kidney and kidney tumor segmentation, a progressive multi-stage cascade dynamically resizes and crops task-specific regions of interest, enabling the model to focus on anatomically relevant structures and improve segmentation accuracy, as shown in Figure 2.

Extensive experiments across multiple prostate and kidney datasets demonstrate that GUSL achieves state-of-the-art performance in prostate and kidney organ segmentation, and competitive results in kidney tumor and mass segmentation. Beyond accuracy, GUSL consistently shows substantial reductions in model size, computational cost (FLOPs), energy consumption, and carbon footprint, highlighting its advantages over conventional deep learning approaches.

These results position [...]

Congratulations to Kevin Yang for passing his Defense! Kevin’s thesis is titled “Interpretable and Efficient Multi-Modal Data Interplay: Algorithms and Applications.” His Dissertation Committee includes Jay Kuo (Chair), Antonio Ortega, and Jesse Thomason (Outside Member). Here is a brief summary of his thesis:

This research addresses the fundamental trade-offs in multimodal learning between interpretability, data efficiency, and computational overhead. As large-scale vision-language models grow increasingly complex, their “black-box” nature and intensive resource requirements present significant barriers to practical deployment. This dissertation introduces four modular frameworks to transition these systems toward explainable, resource-efficient architectures.

First, the Efficient Human-Object Interaction (EHOI) detector decomposes complex interaction tasks into manageable sub-problems, providing transparent intermediate results to support interpretable decision-making. Second, Green Multimodal Alignment (GMA) enhances image-text retrieval by leveraging object detection and semantic clustering, allowing for precise regional interest mapping. Building on these principles, the third work introduces an optimized Video-Text Alignment (VTA) architecture that leverages contrastive learning and specialized data preprocessing to reduce computational costs during inference drastically. Finally, the Semantic and Visual Defect Detector (SVD-Det) bridges the gap between academic research and industrial application. By aligning features across modalities, SVD-Det achieves state-of-the-art performance in AI-generated video detection while maintaining a lightweight structure suitable for real-world use.

Ultimately, these contributions offer a sustainable roadmap for high-performing AI. By prioritizing modularity and transparency, this research establishes an efficient pipeline capable of processing complex, real-world data for both academic inquiry and industrial-scale deployment.