Search

Latent Diffusion Models (LDMs) can generate high-fidelity images from noise, offering a promising approach for augmenting histopathology images for training cancer grading models. While previous works successfully generated high-fidelity histopathology images using LDMs, the generation of image tiles to improve prostate cancer grading has not yet been explored. Additionally, LDMs face challenges in accurately generating admixtures of multiple cancer grades in a tile when conditioned by a tile mask. In this study, we train specific LDMs to generate synthetic tiles that contain multiple Gleason Grades (GGs) by leveraging pixel-wise annotations in input tiles. We introduce a novel framework named Self-Distillation from Separated Conditions (DISC) that generates GG patterns guided by GG masks. Finally, we deploy a training framework for pixel-level and slide-level prostate cancer grading, where synthetic tiles are effectively utilized to improve the cancer grading performance of existing models. As a result, this work surpasses previous works in two domains: 1) our LDMs enhanced with DISC produce more accurate tiles in terms of GG patterns, and 2) our training scheme, incorporating synthetic data, significantly improves the generalization of the baseline model for prostate cancer grading, particularly in challenging cases of rare GG5, demonstrating the potential of generative models to enhance cancer grading when data is limited., Comment: Abstract accepted for ISBI 2024. Extended version to be presented at SynData4CV @ CVPR 2024. See more at https://minhmanho.github.io/disc/

The Frozen Section (FS) technique is a rapid and efficient method, taking only 15-30 minutes to prepare slides for pathologists' evaluation during surgery, enabling immediate decisions on further surgical interventions. However, FS process often introduces artifacts and distortions like folds and ice-crystal effects. In contrast, these artifacts and distortions are absent in the higher-quality formalin-fixed paraffin-embedded (FFPE) slides, which require 2-3 days to prepare. While Generative Adversarial Network (GAN)-based methods have been used to translate FS to FFPE images (F2F), they may leave morphological inaccuracies with remaining FS artifacts or introduce new artifacts, reducing the quality of these translations for clinical assessments. In this study, we benchmark recent generative models, focusing on GANs and Latent Diffusion Models (LDMs), to overcome these limitations. We introduce a novel approach that combines LDMs with Histopathology Pre-Trained Embeddings to enhance restoration of FS images. Our framework leverages LDMs conditioned by both text and pre-trained embeddings to learn meaningful features of FS and FFPE histopathology images. Through diffusion and denoising techniques, our approach not only preserves essential diagnostic attributes like color staining and tissue morphology but also proposes an embedding translation mechanism to better predict the targeted FFPE representation of input FS images. As a result, this work achieves a significant improvement in classification performance, with the Area Under the Curve rising from 81.99% to 94.64%, accompanied by an advantageous CaseFD. This work establishes a new benchmark for FS to FFPE image translation quality, promising enhanced reliability and accuracy in histopathology FS image analysis. Our work is available at https://minhmanho.github.io/f2f_ldm/., Comment: Preprint. Our work is available at https://minhmanho.github.io/f2f_ldm/

We have performed targeted searches of known extragalactic transient events at millimetre wavelengths using nine seasons (2013--2021) of 98, 150, and 229\,GHz Atacama Cosmology Telescope (ACT) observations that mapped ${\sim}40$ per cent of the sky for most of the data volume. Our data cover 88 gamma-ray bursts (GRBs), 12 tidal disruption events (TDEs) and 203 other transients, including supernovae (SNe). We stack our ACT observations to increase the signal-to-noise ratio of the maps. In all cases but one, we do not detect these transients in the ACT data. The single candidate detection (event AT2019ppm), seen at ${\sim}5\sigma$ significance in our data, appears to be due to active galactic nuclei (AGN) activity in the host galaxy coincident with a transient alert. For each source in our search we provide flux upper limits. For example, the medians for the 95 per cent confidence upper limits at 98\,GHz are $15$, $18$, and $16$\,mJy for GRBs, SNe, and TDEs respectively, in the first month after discovery. The projected sensitivity of future wide-area cosmic microwave background (CMB) surveys should be sufficient to detect many of these events using the methods described in this paper., Comment: 14 pages, 7 figures, 8 tables. The full tables of flux upper limits for individual transients are available at https://lambda.gsfc.nasa.gov/product/act/actadv_targeted_transient_constraints_2023_info.html. Accepted for publication by MNRAS

Recently, text-guided image manipulation has received increasing attention in the research field of multimedia processing and computer vision due to its high flexibility and controllability. Its goal is to semantically manipulate parts of an input reference image according to the text descriptions. However, most of the existing works have the following problems: (1) text-irrelevant content cannot always be maintained but randomly changed, (2) the performance of image manipulation still needs to be further improved, (3) only can manipulate descriptive attributes. To solve these problems, we propose a novel image manipulation method that interactively edits an image using complex text instructions. It allows users to not only improve the accuracy of image manipulation but also achieve complex tasks such as enlarging, dwindling, or removing objects and replacing the background with the input image. To make these tasks possible, we apply three strategies. First, the given image is divided into text-relevant content and text-irrelevant content. Only the text-relevant content is manipulated and the text-irrelevant content can be maintained. Second, a super-resolution method is used to enlarge the manipulation region to further improve the operability and to help manipulate the object itself. Third, a user interface is introduced for editing the segmentation map interactively to re-modify the generated image according to the user's desires. Extensive experiments on the Caltech-UCSD Birds-200-2011 (CUB) dataset and Microsoft Common Objects in Context (MS COCO) datasets demonstrate our proposed method can enable interactive, flexible, and accurate image manipulation in real-time. Through qualitative and quantitative evaluations, we show that the proposed model outperforms other state-of-the-art methods., Comment: Accepted to WACV2023

Objectives: The purpose is to apply a previously validated deep learning algorithm to a new thyroid nodule ultrasound image dataset and compare its performances with radiologists. Methods: Prior study presented an algorithm which is able to detect thyroid nodules and then make malignancy classifications with two ultrasound images. A multi-task deep convolutional neural network was trained from 1278 nodules and originally tested with 99 separate nodules. The results were comparable with that of radiologists. The algorithm was further tested with 378 nodules imaged with ultrasound machines from different manufacturers and product types than the training cases. Four experienced radiologists were requested to evaluate the nodules for comparison with deep learning. Results: The Area Under Curve (AUC) of the deep learning algorithm and four radiologists were calculated with parametric, binormal estimation. For the deep learning algorithm, the AUC was 0.69 (95% CI: 0.64 - 0.75). The AUC of radiologists were 0.63 (95% CI: 0.59 - 0.67), 0.66 (95% CI:0.61 - 0.71), 0.65 (95% CI: 0.60 - 0.70), and 0.63 (95%CI: 0.58 - 0.67). Conclusion: In the new testing dataset, the deep learning algorithm achieved similar performances with all four radiologists. The relative performance difference between the algorithm and the radiologists is not significantly affected by the difference of ultrasound scanner., Comment: Clinical Imaging (2023)

We study a mass formula for Artin--Schreier curves of genus $g$ defined over a finite field $k$ of characteristic $p$. For an odd prime $p$ and for small $g$, we determine the number of $k$-isomorphism classes of Artin-Schreier curves of genus $g$, weighted by the order of the centralizer of the Artin-Schreier involution in the automorphism group. This extends earlier results by several authors in characteristic $p=2$. Keywords: Artin-Schreier curve, finite field, automorphism, Mass formula, moduli space, arithmetic statistics

Photonic crystals of close-packed arrays of air spheres in a dielectric background of titania have been fabricated with a novel ceramic technique. Unlike previous methods, ordering of the spheres and the formation of the titania network are performed simultaneously. The photonic crystals exhibit a reflectance peak and a uniform color at the position of the first stop band. The wavelength of the reflectance peak scales very well with the sphere size., Comment: 7 pages, 2 figures

The discovery of the 1144-phase, e.g. CaKFe$_4$As$_4$, creates opportunities to build novel intermetallics with alternative stacking of two parent compounds. Here we formalize the idea by defining a class of bulk crystalline solids with $A$-$B$ stacking (including 1144-phases and beyond), which is a generalization of hetero-structures from few-layer or thin-film semi-conductors to bulk intermetallics. Theoretically, four families of phosphides \textit{AB}(TM)$_4$P$_4$ (TM=Fe, Ru, Co, Ni) are investigated by first-principles calculations, wherein configurational, vibrational and electronic degrees of freedom are considered. It predicts a variety of stable 1144-phases (especially Ru- and Fe-phosphides). Stability rules are found and structural/electronic properties are discussed. Experimentally, we synthesize high-purity CaKRu$_4$P$_4$ as a proof of principle example. The synthetic method is simple and easily applied. Moreover, it alludes to a strategy to explore complex multi-component compounds, facilitated by a phase diagram coordinated by collective descriptors., Comment: 15 pages, 9 figures

Physical photographs now can be conveniently scanned by smartphones and stored forever as a digital version, yet the scanned photos are not restored well. One solution is to train a supervised deep neural network on many digital photos and the corresponding scanned photos. However, it requires a high labor cost, leading to limited training data. Previous works create training pairs by simulating degradation using image processing techniques. Their synthetic images are formed with perfectly scanned photos in latent space. Even so, the real-world degradation in smartphone photo scanning remains unsolved since it is more complicated due to lens defocus, lighting conditions, losing details via printing. Besides, locally structural misalignment still occurs in data due to distorted shapes captured in a 3-D world, reducing restoration performance and the reliability of the quantitative evaluation. To solve these problems, we propose a semi-supervised Deep Photo Scan (DPScan). First, we present a way of producing real-world degradation and provide the DIV2K-SCAN dataset for smartphone-scanned photo restoration. Also, Local Alignment is proposed to reduce the minor misalignment remaining in data. Second, we simulate many different variants of the real-world degradation using low-level image transformation to gain a generalization in smartphone-scanned image properties, then train a degradation network to generalize all styles of degradation and provide pseudo-scanned photos for unscanned images as if they were scanned by a smartphone. Finally, we propose a Semi-Supervised Learning that allows our restoration network to be trained on both scanned and unscanned images, diversifying training image content. As a result, the proposed DPScan quantitatively and qualitatively outperforms its baseline architecture, state-of-the-art academic research, and industrial products in smartphone photo scanning., Comment: Our work is available at https://minhmanho.github.io/dpscan

Molecular dynamics simulations using an interatomic potential developed by artificial neural network deep machine learning are performed to study the local structural order in Al90Tb10 metallic glass. We show that more than 80% of the Tb-centered clusters in Al90Tb10 glass have short-range order (SRO) with their 17 first coordination shell atoms stacked in a '3661' or '15551' sequence. Medium-range order (MRO) in Bergman-type packing extended out to the second and third coordination shells is also clearly observed. Analysis of the network formed by the '3661' and '15551' clusters show that ~82% of such SRO units share their faces or vertexes, while only ~6% of neighboring SRO pairs are interpenetrating. Such a network topology is consistent with the Bergman-type MRO around the Tb-centers. Moreover, crystal structure searches using genetic algorithm and the neural network interatomic potential reveal several low-energy metastable crystalline structures in the composition range close to Al90Tb10. Some of these crystalline structures have the '3661' SRO while others have the '15551' SRO. While the crystalline structures with the '3661' SRO also exhibit the MRO very similar to that observed in the glass, the ones with the '15551' SRO have very different atomic packing in the second and third shells around the Tb centers from that of the Bergman-type MRO observed in the glassy phase.