Refine
Document Type
- Article (8)
- In Proceedings (1)
- ZIB-Report (1)
Language
- English (10)
Has Fulltext
- no (10)
Is part of the Bibliography
- no (10)
Vertebral labelling and segmentation are two fundamental tasks in an automated spine processing pipeline. Reliable and accurate processing of spine images is expected to benefit clinical decision support systems for diagnosis, surgery planning, and population-based analysis of spine and bone health. However, designing automated algorithms for spine processing is challenging predominantly due to considerable variations in anatomy and acquisition protocols and due to a severe shortage of publicly available data. Addressing these limitations, the Large Scale Vertebrae Segmentation Challenge (VerSe) was organised in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2019 and 2020, with a call for algorithms tackling the labelling and segmentation of vertebrae. Two datasets containing a total of 374 multi-detector CT scans from 355 patients were prepared and 4505 vertebrae have individually been annotated at voxel level by a human-machine hybrid algorithm (https://osf.io/nqjyw/, https://osf.io/t98fz/). A total of 25 algorithms were benchmarked on these datasets. In this work, we present the results of this evaluation and further investigate the performance variation at the vertebra level, scan level, and different fields of view. We also evaluate the generalisability of the approaches to an implicit domain shift in data by evaluating the top-performing algorithms of one challenge iteration on data from the other iteration. The principal takeaway from VerSe: the performance of an algorithm in labelling and segmenting a spine scan hinges on its ability to correctly identify vertebrae in cases of rare anatomical variations. The VerSe content and code can be accessed at: https://github.com/anjany/verse.
Background
We assessed the novel MACC1 gene to further stratify stage II colon cancer patients with proficient mismatch repair (pMMR).
Patients and methods
Four cohorts with 596 patients were analyzed: Charité 1 discovery cohort was assayed for MACC1 mRNA expression and MMR in cryo-preserved tumors. Charité 2 comparison cohort was used to translate MACC1 qRT-PCR analyses to FFPE samples. In the BIOGRID 1 training cohort MACC1 mRNA levels were related to MACC1 protein levels from immunohistochemistry in FFPE sections; also analyzed for MMR. Chemotherapy-naïve pMMR patients were stratified by MACC1 mRNA and protein expression to establish risk groups based on recurrence-free survival (RFS). Risk stratification from BIOGRID 1 was confirmed in the BIOGRID 2 validation cohort. Pooled BIOGRID datasets produced a best effect-size estimate.
Results
In BIOGRID 1, using qRT-PCR and immunohistochemistry for MACC1 detection, pMMR/MACC1-low patients had a lower recurrence probability versus pMMR/MACC1-high patients (5-year RFS of 92% and 67% versus 100% and 68%, respectively). In BIOGRID 2, longer RFS was confirmed for pMMR/MACC1-low versus pMMR/MACC1-high patients (5-year RFS of 100% versus 90%, respectively). In the pooled dataset, 6.5% of patients were pMMR/MACC1-low with no disease recurrence, resulting in a 17% higher 5-year RFS (95% CI (12.6-21.3%)) versus pMMR/MACC1-high patients (P=0.037). Outcomes were similar for pMMR/MACC1-low and deficient MMR (dMMR) patients (5-year RFS of 100% and 96%, respectively).
Conclusions
MACC1 expression stratifies colon cancer patients with unfavorable pMMR status. Stage II colon cancer patients with pMMR/MACC1-low tumors have a similar favorable prognosis to those with dMMR with potential implications for the role of adjuvant therapy.
Background
We assessed the novel MACC1 gene to further stratify stage II colon cancer
patients with proficient mismatch repair (pMMR).
Patients and methods
Four cohorts with 596 patients were analyzed: Charité 1 discovery cohort
was assayed for MACC1 mRNA expression and MMR in cryo-preserved
tumors. Charité 2 comparison cohort was used to translate MACC1 qRT-
PCR analyses to FFPE samples. In the BIOGRID 1 training cohort MACC1
mRNA levels were related to MACC1 protein levels from
immunohistochemistry in FFPE sections; also analyzed for MMR.
Chemotherapy-naïve pMMR patients were stratified by MACC1 mRNA and
protein expression to establish risk groups based on recurrence-free
survival (RFS). Risk stratification from BIOGRID 1 was confirmed in the
BIOGRID 2 validation cohort. Pooled BIOGRID datasets produced a best
effect-size estimate.
Results
In BIOGRID 1, using qRT-PCR and immunohistochemistry for MACC1
detection, pMMR/MACC1-low patients had a lower recurrence probability
versus pMMR/MACC1-high patients (5-year RFS of 92% and 67% versus
100% and 68%, respectively). In BIOGRID 2, longer RFS was confirmed
for pMMR/MACC1-low versus pMMR/MACC1-high patients (5-year RFS of
100% versus 90%, respectively). In the pooled dataset, 6.5% of patients
were pMMR/MACC1-low with no disease recurrence, resulting in a 17%
higher 5-year RFS (95% CI (12.6-21.3%)) versus pMMR/MACC1-high
patients (P=0.037). Outcomes were similar for pMMR/MACC1-low and
deficient MMR (dMMR) patients (5-year RFS of 100% and 96%,
respectively).
Conclusions
MACC1 expression stratifies colon cancer patients with unfavorable pMMR
status. Stage II colon cancer patients with pMMR/MACC1-low tumors have
a similar favorable prognosis to those with dMMR with potential
implications for the role of adjuvant therapy.
In this paper, we consider the eigenvalue PDE problem of the infinitesimal generators of metastable diffusion processes. We propose a numerical algorithm based on training artificial neural networks for solving the leading eigenvalues and eigenfunctions of such high-dimensional eigenvalue problem. The algorithm is useful in understanding the dynamical behaviors of metastable processes on large timescales. We demonstrate the capability of our algorithm on a high-dimensional model problem, and on the simple molecular system alanine dipeptide.
We propose Riemannian Denoising Diffusion Probabilistic Models (RDDPMs) for learning distributions on submanifolds of Euclidean space that are level sets of functions, including most of the manifolds relevant to applications. Existing methods for generative modeling on manifolds rely on substantial geometric information such as geodesic curves or eigenfunctions of the Laplace-Beltrami operator and, as a result, they are limited to manifolds where such information is available. In contrast, our method, built on a projection scheme, can be applied to more general manifolds, as it only requires being able to evaluate the value and the first order derivatives of the function that defines the submanifold. We provide a theoretical analysis of our method in the continuous-time limit, which elucidates the connection between our RDDPMs and score-based generative models on manifolds. The capability of our method is demonstrated on datasets from previous studies and on new datasets sampled from two high-dimensional manifolds, i.e. SO(10) and the configuration space of molecular system alanine dipeptide with fixed dihedral angle.
We present EPR-Net, a novel and effective deep learning approach that tackles a crucial challenge in biophysics: constructing potential landscapes for high-dimensional non-equilibrium steady-state (NESS) systems. EPR-Net leverages a nice mathematical fact that the desired negative potential gradient is simply the orthogonal projection of the driving force of the underlying dynamics in a weighted inner-product space. Remarkably, our loss function has an intimate connection with the steady entropy production rate (EPR), enabling simultaneous landscape construction and EPR estimation. We introduce an enhanced learning strategy for systems with small noise, and extend our framework to include dimensionality reduction and state-dependent diffusion coefficient case in a unified fashion. Comparative evaluations on benchmark problems demonstrate the superior accuracy, effectiveness, and robustness of EPR-Net compared to existing methods. We apply our approach to challenging biophysical problems, such as an 8D limit cycle and a 52D multi-stability problem, which provide accurate solutions and interesting insights on constructed landscapes. With its versatility and power, EPR-Net offers a promising solution for diverse landscape construction problems in biophysics.
A connectomic resource for neural cataloguing and circuit dissection of the larval zebrafish brain
(2025)
We present a correlated light and electron microscopy (CLEM) dataset from a 7-day-old larval zebrafish, integrating confocal imaging of genetically labeled excitatory (vglut2a) and inhibitory (gad1b) neurons with nanometer-resolution serial section EM. The dataset spans the brain and anterior spinal cord, capturing >180,000 segmented soma, >40,000 molecularly annotated neurons, and 30 million synapses, most of which were classified as excitatory, inhibitory, or modulatory. To characterize the directional flow of activity across the brain, we leverage the synaptic and cell body annotations to compute region-wise input and output drive indices at single cell resolution. We illustrate the dataset’s utility by dissecting and validating circuits in three distinct systems: water flow direction encoding in the lateral line, recurrent excitation and contralateral inhibition in a hindbrain motion integrator, and functionally relevant targeted long-range projections from a tegmental excitatory nucleus, demonstrating that this resource enables rigorous hypothesis testing as well as exploratory-driven circuit analysis. The dataset is integrated into an open-access platform optimized to facilitate community reconstruction and discovery efforts throughout the larval zebrafish brain.
Euclidean diffusion models have achieved remarkable success in generative modeling across diverse domains, and they have been extended to manifold case in recent advances. Instead of explicitly utilizing the structure of special manifolds as studied in previous works, we investigate direct sampling of the Euclidean diffusion models for general manifold-constrained data in this paper. We reveal the multiscale singularity of the score function in the embedded space of manifold, which hinders the accuracy of diffusion-generated samples. We then present an elaborate theoretical analysis of the singularity structure of the score function by separating it along the tangential and normal directions of the manifold. To mitigate the singularity and improve the sampling accuracy, we propose two novel methods: (1) Niso-DM, which introduces non-isotropic noise along the normal direction to reduce scale discrepancies, and (2) Tango-DM, which trains only the tangential component of the score function using a tangential-only loss function. Numerical experiments demonstrate that our methods achieve superior performance on distributions over various manifolds with complex geometries.