Bacteria tracking and life cycle state classification using graph neural networks and pretrained vision transformers.

Higher-Order Cell Tracking Transformer

Reconstructing lineages from live-imaging microscopy requires linking cell detections across time, including through cell divisions. A common approach is to construct a candidate graph and associate cell segmentations (nodes) across frames. However, these and other existing methods overlook two structural obstacles in candidate tracking graphs: (i) cell divisions entangle distinct lineage paths in the node embedding space, and (ii) edges sharing a node have near-random label agreement, so the candidate-graph topology carries no useful information for graph neural networks to aggregate. We propose the \textbf{Higher-Order Cell Tracking Transformer} (HOCT), an edge-centric architecture in which candidate cell links attend to one another under a 3D geometric prior, resolving both issues. Evaluated on the Cell Tracking Challenge and a bacteria division benchmark, HOCT achieves state-of-the-art results without deep pre-trained image encoders. Moreover, the proposed approach is easier to fine-tune, quickly reducing tracking errors by 59% with 400 annotations in a human-in-the-loop setting, outperforming LoRA fine-tuning of competing transformer baselines (6.75% improvement).

Temporal dynamics improves machine learning-based prediction of cell state from quantitative phase imaging

Cell morphology reflects cell health and can distinguish cell-cycle stage, growth arrest, and distinct pathways of cell death. Live, label-free quantitative phase imaging (QPI) captures these features non-invasively and with high temporal resolution, yet many image-based classifiers rely on single frames and cannot separate states whose differences emerge only over time. How much temporal information is needed, and which architecture best exploits it, remain open questions. We assembled 1,874 QPI timelapse sequences spanning six cell states (interphase, mitosis, cell cycle arrest, apoptosis, ferroptosis, and necroptosis) and compared two-dimensional convolutional neural networks (CNNs) with a three-dimensional (3D) spatiotemporal CNN across increasing frame counts. Accuracy improved as frames were added, with the largest gain between one and three frames. The 2D models saturated beyond three frames, whereas the 3D architecture kept improving, reaching 96.5% accuracy and a 3.5% error rate at eleven frames. The temporal information needed tracked the timescale of each process: mitosis was resolved from a single frame, while ferroptosis benefited most from extended sequences. Overall, these results show that dynamic information, rather than static morphology alone, drives accurate cell-state classification, and that 3D architectures are needed to fully exploit it for label-free dynamic phenotyping.

Bachelor Computer Science & Bioinformatics Improving Early S Phase Detection in Hoechst-Based Cell-Cycle Segmentation Using Temporal Context

Standardization, Benchmarking, and Fine-Tuning in Deep Learning-Based Cell Segmentation and Tracking

Bioimage analysis workflows for living cells typically involve a sequence of steps, including image acquisition, preprocessing, cell segmentation, object classification, and temporal tracking. Cell segmentation aims to identify and separate individual cells in microscopy images, whereas cell tracking links these segmented objects across time-lapse image sequences to quantify cell movement, morphology, and dynamic behavior. These tasks are challenging because microscopy datasets often vary in image quality and may contain imaging noise, variable staining, overlapping cells, heterogeneous cell morphologies, and changing acquisition conditions. Over the past decade, the field has shifted from heuristic and rule-based image processing toward deep learning-based approaches, supported by advances in convolutional neural networks, foundation models, and increasingly standardized datasets. This transition has been closely connected to the development of interoperable data formats, shared benchmarks, and open-source bioimage analysis ecosystems. The present review discusses this evolution with a focus on standardization, benchmarking, and data-centric model adaptation strategies. In this context, data-centric strategies refer to approaches that improve model performance primarily through better data selection, annotation, fine-tuning, and expert feedback rather than through architecture design alone. Particular attention is given to small-sample fine-tuning and human-in-the-loop workflows, which aim to adapt pretrained segmentation and tracking models to laboratory-specific microscopy data while reducing annotation effort and improving the reliability of downstream biological conclusions.

Automated Deep Learning Protocol for Quantifying Cell Viability in Live/Dead Fluorescence Microscopy.

Unsupervised Learning of Cell Instances with Generative Routing Pyramids

Identifying and representing object instances such as cells or nuclei is a common task in microscopy image analysis. Established machine learning workflows typically use supervised detection or segmentation followed by feature extraction or classification, which requires manual annotations and treats instance segmentation and cell representation as separate stages. We describe a new unsupervised method for cell instance segmentation and phenotypic classification from unlabeled microscopy images. Our method is based on reconstructing each image using a coarse-to-fine routing pyramid that associates pixels with spatially sparse latent sources. The resulting pixel-to-latent associations yield instance masks, while the source latents encode cell morphology. We demonstrate competitive performance in instance segmentation across diverse cell morphologies and imaging modalities, as well as generative modeling of cellular phenotypes under perturbations. Source code and checkpoints are available at https://github.com/weigertlab/routing-pyramids.