Biology subjects

Pound, M. P.

Publications and source records attributed to Pound, M. P..

6 recordsLinked to original sources

bioRxiv2026

Diffusion-Based Synthetic Pretraining for Root Segmentation in Rhizotron Images

Rhizotrons allow root system architecture to be studied non-destructively in soil. However, automated analysis of rhizotron images remains difficult: roots have low contrast against soil, soil particles resemble fine roots, and deep learning models need many manually annotated images that are slow and subjective to produce. Synthetic images could provide paired images and annotations at scale, but rhizotron images are largely absent from the data used to train generative models. We built a pipeline combining Low-Rank Adaptation (LoRA), ControlNet and inpainting to generate synthetic rhizotron images with paired root masks. We compared training on synthetic data only, real data only, and synthetic pretraining followed by fine-tuning on real images, using three segmentation models: nnU-Net v2, YOLO26 and Mask2Former. Synthetic-only training reached a Dice score of up to 0.60. Synthetic pretraining followed by fine-tuning on the full real training set (two annotated rhizotron images) reached 0.80, and nnU-Net v2 and Mask2Former reached 0.62 with only two annotated image patches containing visible roots. When real annotations were limited, pretraining also improved segmentation of unseen crops, including lettuce, barley, wheat, maize, tomato, cucumber, French bean and field bean. We also introduced a Soft Dice score that compares predictions with probabilistic consensus maps from seven independent annotators, accounting for annotator disagreement. Models matched average human agreement on low-ambiguity patches but remained below it on high-ambiguity patches. In conclusion, diffusion-generated images can complement real annotations and help scale root segmentation, although their effect on downstream root trait measurements still needs to be validated.

plant biology↗

bioRxiv2019

Three Dimensional Root CT Segmentation using Multi-Resolution Encoder-Decoder Networks

We address the complex problem of reliably segmenting root structure from soil in X-ray Computed Tomography (CT) images. We utilise a deep learning approach, and propose a state-of-the-art multi-resolution architecture based on encoder-decoders. While previous work in encoder-decoders implies the use of multiple resolutions simply by downsampling and upsampling images, we make this process explicit, with branches of the network tasked separately with obtaining local high-resolution segmentation, and wider low-resolution contextual information. The complete network is a memory efficient implementation that is still able to resolve small root detail in large volumetric images. We evaluate our approach by comparing against a number of different encoder-decoder based architectures from the literature, as well as a popular existing image analysis tool designed for root CT segmentation. We show qualitatively and quantitatively that a multi-resolution approach offers substantial accuracy improvements over a both a small receptive field size in a deep network, or a larger receptive field in a shallower network. We obtain a Dice score of 0.59 compared with 0.41 for the closest competing method. We then further improve performance using an incremental learning approach, in which failures in the original network are used to generate harder negative training examples. Results of this process raise the precision of the network, and improve the Dice score to 0.66. Our proposed method requires no user interaction, is fully automatic, and identifies large and fine root material throughout the whole volume. The 3D segmented output of our method is well-connected, allowing the recovery of structured representations of root system architecture, and so may be successfully utilised in root phenotyping.

bioinformatics↗

bioRxiv2019

Volumetric Segmentation of Cell Cycle Markers in Confocal Images

I.AO_SCPCAPBSTRACTC_SCPCAPUnderstanding plant growth processes is important for many aspects of biology and food security. Automating the observations of plant development - a process referred to as plant phenotyping - is increasingly important in the plant sciences, and is often a bottleneck. Automated tools are required to analyse the data in images depicting plant growth. In this paper, a deep learning approach is developed to locate fluorescent markers in 3D timeseries microscopy images. The approach is not dependant on marker morphology; only simple 3D point location annotations are required for training. The approach is evaluated on an unseen timeseries comprising several volumes, capturing growth of plants. Results are encouraging, with an average recall of 0.97 and average F-score of 0.78, despite only a very limited number of simple training annotations. In addition, an in-depth analysis of appropriate loss functions is conducted. To accompany [the finally-published] paper we are releasing the 4D point annotation tool used to generate the annotations, in the form of a plugin for the popular ImageJ (Fiji) software. Network models will be released online.

plant biology↗

bioRxiv2019

RootNav 2.0: Deep Learning for Automatic Navigation of Complex Plant Root Architectures

We present a new image analysis approach that provides fully-automatic extraction of complex root system architectures from a range of plant species in varied imaging setups. Driven by modern deep-learning approaches, RootNav 2.0 replaces previously manual and semi-automatic feature extraction with an extremely deep multi-task Convolutional Neural Network architecture. The network has been designed to explicitly combine local pixel information with global scene information in order to accurately segment small root features across high-resolution images. In addition, the network simultaneously locates seeds, and first and second order root tips to drive a search algorithm seeking optimal paths throughout the image, extracting accurate architectures without user interaction. The proposed method is evaluated on images of wheat (Triticum aestivum L.) from a seedling assay. The results are compared with semi-automatic analysis via the original RootNav tool, demonstrating comparable accuracy, with a 10-fold increase in speed. We then demonstrate the ability of the network to adapt to different plant species via transfer learning, offering similar accuracy when transferred to an Arabidopsis thaliana plate assay. We transfer for a final time to images of Brassica napus from a hydroponic assay, and still demonstrate good accuracy despite many fewer training images. The tool outputs root architectures in the widely accepted RSML standard, for which numerous analysis packages exist (http://rootsystemml.github.io/), as well as segmentation masks compatible with other automated measurement tools.

bioinformatics↗

bioRxiv2019

Identification of nitrogen-dependent QTL and underlying genes for root system architecture in hexaploid wheat

The root system architecture (RSA) of a crop has a profound effect on the uptake of nutrients and consequently the potential yield. However, little is known about the genetic basis of RSA and resource adaptive responses in wheat (Triticum aestivum L.). Here, a high-throughput germination paper plant phenotyping system was used to identify seedling traits in a wheat doubled haploid mapping population, Savannah x Rialto. Significant genotypic and nitrate-N treatment variation was found across the population for seedling traits with distinct trait grouping for root size-related traits and root distribution-related traits. Quantitative trait locus (QTL) analysis identified a total of 59 seedling trait QTLs. Across two nitrate treatments, 27 root QTLs were specific to the nitrate treatment. Transcriptomic analyses for one of the QTLs on chromosome 2D found under low nitrate conditions was pursued revealing gene enrichment in N-related biological processes and 17 candidate up-regulated genes with possible involvement in a root angle response. Together, these findings provide genetic insight into root system architecture and plant adaptive responses to nitrate and provide targets that could help improve N capture in wheat.

plant biology↗

bioRxiv2017

Deep Learning for Multi-task Plant Phenotyping

Plant phenotyping has continued to pose a challenge to computer vision for many years. There is a particular demand to accurately quantify images of crops, and the natural variability and structure of these plants presents unique difficulties. Recently, machine learning approaches have shown impressive results in many areas of computer vision, but these rely on large datasets that are at present not available for crops. We present a new dataset, called ACID, that provides hundreds of accurately annotated images of wheat spikes and spikelets, along with image level class annotation. We then present a deep learning approach capable of accurately localising wheat spikes and spikelets, despite the varied nature of this dataset. As well as locating features, our network offers near perfect counting accuracy for spikes (95.91%) and spikelets (99.66%). We also extend the network to perform simultaneous classification of images, demonstrating the power of multi-task deep architectures for plant phenotyping. We hope that our dataset will be useful to researchers in continued improvement of plant and crop phenotyping. With this in mind, alongside the dataset we will make all code and trained models available online.

plant biology↗