arXiv subjects

Shipra Jain

Publications and source records attributed to Shipra Jain.

4 recordsLinked to original sources

arXiv2026

Teeth2Point: A Two-Stage Dental CBCT ROI-to-Point Segmentation Framework

Modern deep learning architectures have demonstrated strong performance in dental CBCT segmentation. One remaining crucial challenge is accurate tooth labeling in cases with missing or malpositioned teeth, which are highly relevant for dental practice. Transformer-based architectures should in theory be able to resolve such ambiguities using global anatomical context. However, due to the high resolution of CBCT volumes and the wide spatial distribution of teeth within volumes, dense patch-based volumetric processing faces an inherent trade-off. Computational costs limit the number of patches that can be used in self-attention and thus, one can either increase the extent of the context captured in self-attention or capture fine-grained structural details by using small patches, but not both. In this work, we present Teeth2Point, an efficient point-based transformer framework for dental CBCT semantic segmentation that can avoid this trade-off. Teeth2Point first localizes volumetric regions of interest (ROIs) surrounding teeth using a convolutional model, then converts ROIs into point tokens using adaptive sampling. A transformer model predicts accurate segmentations using the point tokens, which allow capturing global context while retaining high resolution. The transformer is first pretrained using self-supervised learning (SSL), in the style of DINO but using domain-specific augmentation strategies, followed by supervised finetuning. The SSL pretraining, which includes random token masking, provides robustness to complex anatomical variations. Compared with the strongest two-stage baseline, Teeth2Point improves abnormal-case performance by 1.44 DSC points on average across four datasets; relative to the first-stage nnU-Net, the gain is 1.9 points.

cs.CV↗

arXiv2026

SA-VIS: Sparse frame Annotations for training Video Instance Segmentation

Recent online video instance segmentation (VIS) methods have achieved impressive results, thus becoming the preferred approach to segment instances in videos. Despite the resurgence of impressive single image models, the online (or semi-online) VIS approaches outperform single-image models (e.g., based on SAM) by using long sequences of densely annotated frames during training. However,such a training setup of VIS is expensive in the sense of compute as well as dense annotations required. In order to solve these major flaws, we argue that the effective modeling of the instances and their evolution in videos do not require densely annotated frames. To that end, we propose a simple and effective module, called Past-frames Feature Propagation (PFP) which aggregates low-dimensional features from the image encoder of multiple frames. This simple low-compute module provides tremendous learning capability in using sparse video frame labels for end-to-end training. Combined with a light-weight frame-specific Instance Queries, our Sparse frame Annotation VIS (SA-VIS) significantly improves performance over its baseline. Most interestingly, our simple design that avoids complexities effectively bridges the gap in accuracy between training on sparsely and densely annotated video sequences. This translates to a mere 0.4% drop in performance of SA-VIS when using annotations for only 1/5 of the images in the dataset. Empirically, SA-VIS shows strong improvements over the baseline on YouTube-VIS 2019/2021/2022 and Occluded VIS (OVIS) and an over 1% improvement in AP on the state-of-the-art in a limited annotations scenario.

cs.CV↗

arXiv2025

In silico tool for identification of colorectal cancer from cell-free DNA biomarkers

Colorectal cancer remains a major global health concern, with early detection being pivotal for improving patient outcomes. In this study, we leveraged high throughput methylation profiling of cellfree DNA to identify and validate diagnostic biomarkers for CRC. The GSE124600 study data were downloaded from the Gene Expression Omnibus, as the discovery cohort, comprising 142 CRC and 132 normal cfDNA methylation profiles obtained via MCTA seq. After preprocessing and filtering, 97,863 CpG sites were retained for further analysis. Differential methylation analysis using statistical tests identified 30,791 CpG sites as significantly altered in CRC samples, where p is less than 0.05. Univariate scoring enabled the selection of top ranking features, which were further refined using multiple feature selection algorithms, including Recursive Feature Elimination, Sequential Feature Selection, and SVC L1. Various machine learning models such as Logistic Regression, Support Vector Machines, Random Forest, and Multi layer Perceptron were trained and tested using independent validation datasets. The best performance was achieved with an MLP model trained on 25 features selected by RFE, reaching an AUROC of 0.89 and MCC of 0.78 on validation data. Additionally, a deep learning based convolutional neural network achieved an AUROC of 0.78. Functional annotation of the most predictive CpG sites identified several genes involved in key cellular processes, some of which were validated for differential expression in CRC using the GEPIA2 platform. Our study highlights the potential of cfDNA methylation markers combined with ML and DL models for noninvasive and accurate CRC detection, paving the way for clinically relevant diagnostic tools.

q-bio.GN↗

arXiv2021

Scaling Semantic Segmentation Beyond 1K Classes on a Single GPU

The state-of-the-art object detection and image classification methods can perform impressively on more than 9k and 10k classes, respectively. In contrast, the number of classes in semantic segmentation datasets is relatively limited. This is not surprising when the restrictions caused by the lack of labeled data and high computation demand for segmentation are considered. In this paper, we propose a novel training methodology to train and scale the existing semantic segmentation models for a large number of semantic classes without increasing the memory overhead. In our embedding-based scalable segmentation approach, we reduce the space complexity of the segmentation model's output from O(C) to O(1), propose an approximation method for ground-truth class probability, and use it to compute cross-entropy loss. The proposed approach is general and can be adopted by any state-of-the-art segmentation model to gracefully scale it for any number of semantic classes with only one GPU. Our approach achieves similar, and in some cases, even better mIoU for Cityscapes, Pascal VOC, ADE20k, COCO-Stuff10k datasets when adopted to DeeplabV3+ model with different backbones. We demonstrate a clear benefit of our approach on a dataset with 1284 classes, bootstrapped from LVIS and COCO annotations, with three times better mIoU than the DeeplabV3+ model.

cs.CV↗