Jie Tian’s research while affiliated with University of Science and Technology of China and other places

A core challenge in survival analysis is to model the distribution of time-to-event data, where the event of interest may be a death, failure, or occurrence of a specific event. Previous studies have showed that ranking and maximum likelihood estimation loss functions are widely-used learning approaches for survival analysis. However, ranking loss only focus on the ranking of survival time and does not consider potential effect of samplesâ exact survival time values. Furthermore, the maximum likelihood estimation is unbounded and easily subject to outliers (e.g., censored data), which may cause poor performance of modeling. To handle the complexities of learning process and exploit valuable survival time values, we propose a time-adaptive coordinate loss function, TripleSurv, to achieve adaptive adjustments by introducing the differences in the survival time between sample pairs into the ranking, which can encourage the model to quantitatively rank relative risk of pairs, ultimately enhancing the accuracy of predictions. Most importantly, the TripleSurv is proficient in quantifying the relative risk between samples by ranking ordering of pairs, and consider the time interval as a trade-off to calibrate the robustness of model over sample distribution. Our TripleSurv is evaluated on three real-world survival datasets and a public synthetic dataset. The results show that our method outperforms the state-of-the-art methods and exhibits good model performance and robustness on modeling various sophisticated data distributions with different censor rates. Our code is available at upon acceptance.

The second near-infrared (NIR-II) fluorescence imaging has become a new imaging mode due to its characteristics of real-time intraoperative imaging. The NIR-IIb window (1500–1700 nm) has stronger light penetration and has a clearer imaging effect than the NIR-IIa window (1000–1300 nm). The molecular probe currently used for human imaging (indocyanine green) can only be used for NIR-IIa window imaging, and there is a lack of effective molecular probes for NIR-IIb imaging. In addition, there are various types of NIR-II fluorescence images, and it is difficult for a single neural network model to enhance multiple types of images. To solve these problems, we design a universal NIR-II fluorescence image enhancement model to transfer the style of NIR-IIb images to NIR-IIa images and improve the quality of fluorescence images. The model is based on an encoder-decoder framework and can process multiple types of NIR-II fluorescence images simultaneously. Specifically, it includes a module for fusion of low-quality and high-quality image feature maps, the covariance weighted attention network (CWANet), which improves the attention mechanism through covariance weight, allowing the model to filter the style features that are irrelevant or detrimental to image structure in the attention mechanism and pay attention to the key content details of the image. Furthermore, we propose a multiple high-frequency matching loss, which matches the high-frequency information of the enhanced image with the NIR-IIa image to further maintain the structure. Extensive experiments demonstrate that our model generates results that exceed state-of-the-art models, with impressive image enhancement effect.

3D reconstruction from multi-view images is considered as a longstanding problem in computer vision and graphics. In order to achieve high-fidelity geometry and appearance of 3D scenes, this paper proposes a novel geometric object learning method for multi-view reconstruction with fuzzy set theory . We establish a new neural 3D reconstruction theoretical frame called neural fuzzy geometric representation (NeuFG), which is a special type of implicit representation of geometric scene that only takes value in [0, 1]. NeuFG is essentially a volume image, and thus can be visualized directly with the conventional volume rendering technique. Extensive experiments on two public datasets, i.e., DTU and BlendedMVS, show that our method has the ability of accurately reconstructing complex shapes with vivid geometric details, without the requirement of mask supervision. Both qualitative and quantitative comparisons demonstrate that the proposed method has superior performance over the state-of-the-art neural scene representation methods. The code will be released on GitHub soon.

Prognostic assessment remains a critical challenge in medical research, often limited by the lack of well-labeled data. In this work, we introduce ContraSurv, a weakly-supervised learning framework based on contrastive learning, designed to enhance prognostic predictions in 3D medical images. ContraSurv utilizes both the self-supervised information inherent in unlabeled data and the weakly-supervised cues present in censored data, refining its capacity to extract prognostic representations. For this purpose, we establish a Vision Transformer architecture optimized for our medical image datasets and introduce novel methodologies for both self-supervised and supervised contrastive learning for prognostic assessment. Additionally, we propose a specialized supervised contrastive loss function and introduce SurvMix, a novel data augmentation technique for survival analysis. Evaluations were conducted across three cancer types and two imaging modalities on three real-world datasets. The results confirmed the enhanced performance of ContraSurv over competing methods, particularly in data with a high censoring rate.

Magnetic particle imaging (MPI) has emerged as a novel technology utilizing superparamagnetic nanoparticles as tracers, essential for disease diagnosis and treatment guidance in preclinical animal models. Unlike other modalities, MPI provides high sensitivity, deep tissue penetration, and no signal attenuation. However, existing MPI tracers suffer from “always‐on” signals, which complicate organ‐specific imaging and hinder accuracy. To overcome these challenges, we have developed a responsive MPI tracer using pH‐responsive PdFe alloy particles coated with a gatekeeper polymer. This tracer exhibits pH‐sensitive Fe release and modulation of the MPI signal, enabling selective imaging with a higher signal‐to‐noise ratio and intratumoral pH quantification. Notably, this responsive tracer facilitates subtraction‐enhanced MPI imaging, effectively eliminating interference from liver uptake and expanding the scope of abdominal imaging. Additionally, the tracer employs a dual‐function mechanism for adaptive cancer therapy, combining pH‐switchable enzyme‐like catalysis with dual‐key co‐activation of ROS generation, and Pd skeleton that scavenges free radicals to minimize Fe‐related toxicity. This advancement promises to significantly expand MPI's applicability in diagnostics and therapeutic monitoring, marking a leap forward in imaging technology.