Search

Mounting observational evidence indicates that cold circumstellar gas is present in debris disk systems. This work focuses on various dynamical processes that debris-disk gas may undergo. We review five mechanisms that can transport angular momentum and their applications to debris disks. These include molecular viscosity, hydrodynamic turbulence, magnetohydrodynamic turbulence, magnetized disk winds, and laminar magnetic stress. We find that molecular viscosity can result in $\alpha$ as high as $\lesssim 0.1$ for sufficiently low densities, while the Rossby wave instability is a possible source of hydrodynamic turbulence and structure formation. We argue that the vertical shear instability is unlikely due to the long cooling times. The onset of the magnetorotational instability (MRI) is dichotomous: for low density disks the MRI can be excited at the midplane, while for high mass disks it may only be operating at $z>2-3H$, if at all. The MHD wind and laminar magnetic stress mechanisms rely on the configuration and strength of any background large-scale magnetic field, the existence of which is uncertain and possibly unlikely. We conclude that the dominant mechanism and its efficiency in transporting angular momentum varies from one system to the other, depending especially closely on the gas density. More detailed analyses shall be performed in the future focusing on representative, nearby debris disks., Comment: 16 pages, 9 figures, submitted to MNRAS and revised

Multi-task robot learning holds significant importance in tackling diverse and complex scenarios. However, current approaches are hindered by performance issues and difficulties in collecting training datasets. In this paper, we propose GeRM (Generalist Robotic Model). We utilize offline reinforcement learning to optimize data utilization strategies to learn from both demonstrations and sub-optimal data, thus surpassing the limitations of human demonstrations. Thereafter, we employ a transformer-based VLA network to process multi-modal inputs and output actions. By introducing the Mixture-of-Experts structure, GeRM allows faster inference speed with higher whole model capacity, and thus resolves the issue of limited RL parameters, enhancing model performance in multi-task learning while controlling computational costs. Through a series of experiments, we demonstrate that GeRM outperforms other methods across all tasks, while also validating its efficiency in both training and inference processes. Additionally, we uncover its potential to acquire emergent skills. Additionally, we contribute the QUARD-Auto dataset, collected automatically to support our training approach and foster advancements in multi-task quadruped robot learning. This work presents a new paradigm for reducing the cost of collecting robot data and driving progress in the multi-task learning community. You can reach our project and video through the link: https://songwxuan.github.io/GeRM/ .

Past studies on end-to-end meeting transcription have focused on model architecture and have mostly been evaluated on simulated meeting data. We present a novel study aiming to optimize the use of a Speaker-Attributed ASR (SA-ASR) system in real-life scenarios, such as the AMI meeting corpus, for improved speaker assignment of speech segments. First, we propose a pipeline tailored to real-life applications involving Voice Activity Detection (VAD), Speaker Diarization (SD), and SA-ASR. Second, we advocate using VAD output segments to fine-tune the SA-ASR model, considering that it is also applied to VAD segments during test, and show that this results in a relative reduction of Speaker Error Rate (SER) up to 28%. Finally, we explore strategies to enhance the extraction of the speaker embedding templates used as inputs by the SA-ASR system. We show that extracting them from SD output rather than annotated speaker segments results in a relative SER reduction up to 20%., Comment: Submitted to Odyssey 2024

Understanding the anatomy of renal pathology is crucial for advancing disease diagnostics, treatment evaluation, and clinical research. The complex kidney system comprises various components across multiple levels, including regions (cortex, medulla), functional units (glomeruli, tubules), and cells (podocytes, mesangial cells in glomerulus). Prior studies have predominantly overlooked the intricate spatial interrelations among objects from clinical knowledge. In this research, we introduce a novel universal proposition learning approach, called panoramic renal pathology segmentation (PrPSeg), designed to segment comprehensively panoramic structures within kidney by integrating extensive knowledge of kidney anatomy. In this paper, we propose (1) the design of a comprehensive universal proposition matrix for renal pathology, facilitating the incorporation of classification and spatial relationships into the segmentation process; (2) a token-based dynamic head single network architecture, with the improvement of the partial label image segmentation and capability for future data enlargement; and (3) an anatomy loss function, quantifying the inter-object relationships across the kidney., Comment: IEEE / CVF Computer Vision and Pattern Recognition Conference 2024

Understanding the way cells communicate, co-locate, and interrelate is essential to understanding human physiology. Hematoxylin and eosin (H&E) staining is ubiquitously available both for clinical studies and research. The Colon Nucleus Identification and Classification (CoNIC) Challenge has recently innovated on robust artificial intelligence labeling of six cell types on H&E stains of the colon. However, this is a very small fraction of the number of potential cell classification types. Specifically, the CoNIC Challenge is unable to classify epithelial subtypes (progenitor, endocrine, goblet), lymphocyte subtypes (B, helper T, cytotoxic T), or connective subtypes (fibroblasts, stromal). In this paper, we propose to use inter-modality learning to label previously un-labelable cell types on virtual H&E. We leveraged multiplexed immunofluorescence (MxIF) histology imaging to identify 14 subclasses of cell types. We performed style transfer to synthesize virtual H&E from MxIF and transferred the higher density labels from MxIF to these virtual H&E images. We then evaluated the efficacy of learning in this approach. We identified helper T and progenitor nuclei with positive predictive values of $0.34 \pm 0.15$ (prevalence $0.03 \pm 0.01$) and $0.47 \pm 0.1$ (prevalence $0.07 \pm 0.02$) respectively on virtual H&E. This approach represents a promising step towards automating annotation in digital pathology.

Domain wall (DW) devices have garnered recent interest for diverse applications including memory, logic, and neuromorphic primitives; fast, accurate device models are therefore imperative for large-scale system design and verification. Extant DW motion models are sub-optimal for large-scale system design either over-consuming compute resources with physics-heavy equations or oversimplifying the physics, drastically reducing model accuracy. We propose a DW model inspired by the phenomenological similarities between motions of a DW and a classical object being acted on by forces like air resistance or static friction. Our proposed phenomenological model predicts DW motion within 1.2% on average compared with micromagnetic simulations that are 400 times slower. Additionally our model is seven times faster than extant collective coordinate models and 14 times more accurate than extant hyper-reduced models making it an essential tool for large-scale DW circuit design and simulation. The model is publicly posted along with scripts that automatically extract model parameters from user-provided simulation or experimental data to extend the model to alternative micromagnetic parameters.

Multi-modal learning adeptly integrates visual and textual data, but its application to histopathology image and text analysis remains challenging, particularly with large, high-resolution images like gigapixel Whole Slide Images (WSIs). Current methods typically rely on manual region labeling or multi-stage learning to assemble local representations (e.g., patch-level) into global features (e.g., slide-level). However, there is no effective way to integrate multi-scale image representations with text data in a seamless end-to-end process. In this study, we introduce Multi-Level Text-Guided Representation End-to-End Learning (mTREE). This novel text-guided approach effectively captures multi-scale WSI representations by utilizing information from accompanying textual pathology information. mTREE innovatively combines - the localization of key areas (global-to-local) and the development of a WSI-level image-text representation (local-to-global) - into a unified, end-to-end learning framework. In this model, textual information serves a dual purpose: firstly, functioning as an attention map to accurately identify key areas, and secondly, acting as a conduit for integrating textual features into the comprehensive representation of the image. Our study demonstrates the effectiveness of mTREE through quantitative analyses in two image-related tasks: classification and survival prediction, showcasing its remarkable superiority over baselines.

Safety and robustness are crucial factors in developing trustworthy autonomous vehicles. One essential aspect of addressing these factors is to equip vehicles with the capability to predict future trajectories for all moving objects in the surroundings and quantify prediction uncertainties. In this paper, we propose the Sequential Neural Variational Agent (SeNeVA), a generative model that describes the distribution of future trajectories for a single moving object. Our approach can distinguish Out-of-Distribution data while quantifying uncertainty and achieving competitive performance compared to state-of-the-art methods on the Argoverse 2 and INTERACTION datasets. Specifically, a 0.446 meters minimum Final Displacement Error, a 0.203 meters minimum Average Displacement Error, and a 5.35% Miss Rate are achieved on the INTERACTION test set. Extensive qualitative and quantitative analysis is also provided to evaluate the proposed model. Our open-source code is available at https://github.com/PurdueDigitalTwin/seneva., Comment: Accepted at CVPR 2024

The rapid advancement of next-generation wireless communication systems, autonomous driving technologies, and the Internet of Things (IoT) has created an increasing demand for high data rates, lower latency, and improved network capacity. To meet these requirements, the sixth-generation (6G) wireless communication networks are moving towards frequencies above 100 GHz, where a broader spectrum can be utilized, facilitating the effective integration and functioning of sophisticated communication systems and IoT applications. Developing transceivers for sub-terahertz (sub-THz) frequencies requires the optimization of radio frequency (RF) front-end components, which play a crucial role in determining the overall system performance. Power amplifiers, in particular, are essential in the transmitter because they are responsible for amplifying the signal prior to transmission. Moreover, achieving high output power across an extensive system bandwidth is necessary for efficient signal transmission. Additionally, the power amplifier is among the most power-consuming components within a transceiver system, making the minimization of power consumption highly desirable.In this dissertation, the designs and implementations of wideband power amplifiers for sub THz frequencies are explored. The wideband power amplifiers utilizing an Indium Phosphide (InP) heterojunction bipolar transistor (HBT) process are presented, which employ a common base double-stack (CB 2-stack) with a split-top topology to achieve high gain and output power across a wide bandwidth. One power amplifier demonstrates an average gain of 16 dB and a maximum 3-dB bandwidth of 73 GHz, ranging from 130 to 203 GHz, with the measured saturated output power (Psat) higher than 10 dBm up to 170 GHz. In order to attain increased gain and output power, the second design incorporates both a cascade topology and power combiner into the pre-existing common base PA. The simulation achieved a gain of 23 dB and Psat ex

International efforts to avoid dangerous climate change have historically focused on reducing energy-related CO2 emissions from countries with either the largest economies (e.g. the EU and the USA) and/or the largest populations (e.g. China and India). However, in recent years, emissions have surged among a different and much less-examined group of countries, raising concerns that a next generation of high-emitting economies will obviate current mitigation targets. Here, we analyse the trends and drivers of emissions in each of the 59 countries where emissions in 2010-2018 grew faster than the global average (excluding China and India), project their emissions under a range of longer-term energy scenarios and estimate the costs of decarbonization pathways. Total emissions from these 'emerging emitters' reach as much as 7.5 GtCO2/year in the baseline 2.5° scenario-substantially greater than the emissions from these regions in previously published scenarios that would limit warming to 1.5°C or even 2°C. Such unanticipated emissions would in turn require non-emitting energy deployment from all sectors within these emerging emitters, and faster and deeper reductions in emissions from other countries to meet international climate goals. Moreover, the annual costs of keeping emissions at the low level are in many cases 0.2%-4.1% of countries' gross domestic production, pointing to potential trade-offs with poverty-reduction goals and/or the need for economic support and low-carbon technology transfer from historically high-emitting countries. Our results thus highlight the critical importance of ramping up mitigation efforts in countries that to this point have been largely ignored.

Currently, there are four mainstream taxonomic opinions on the classification of the class Halobacteria at the family and order levels. The International Committee on Systematics of Prokaryotes Subcommittee on the Taxonomy of Halobacteria (ICSP), List of Prokaryotic names with Standing in Nomenclature (LPSN) and National Centre for Biotechnology Information (NCBI) adopted taxonomies have three to four orders and up to eight families, while the Genome Taxonomy Database (GTDB) taxonomy proposes only one order with nine families. To resolve the taxonomic inconsistency, phylogenomic analyses based on concatenated single-copy orthologous proteins and 122 concatenated conserved single-copy marker proteins were conducted to infer the taxonomic status of the current representatives of the class Halobacteria at the family and order levels. The current 76 genera with validly published names of the class Halobacteria were able to be assigned into eight families in one order. On the basis of these results, it is proposed that the current species with validly published names of the class Halobacteria should be remerged into the order Halobacteriales, then assigned to eight families, Haladaptataceae, Haloarculaceae, Halobacteriaceae, Halococcaceae, Haloferacaceae, Natronoarchaeaceae, Natrialbaceae and Halorubellaceae. Thus, Haladaptataceae fam. nov. is described based on Haladaptatus, Halomicrococcus and Halorussus and Halorubellaceae fam. nov. is proposed incorporating Haloarchaeobius and Halorubellus, respectively.

Currently, there are four mainstream taxonomic opinions on the classification of the class Halobacteria at the family and order levels. The International Committee on Systematics of Prokaryotes Subcommittee on the Taxonomy of Halobacteria (ICSP), List of Prokaryotic names with Standing in Nomenclature (LPSN) and National Centre for Biotechnology Information (NCBI) adopted taxonomies have three to four orders and up to eight families, while the Genome Taxonomy Database (GTDB) taxonomy proposes only one order with nine families. To resolve the taxonomic inconsistency, phylogenomic analyses based on concatenated single-copy orthologous proteins and 122 concatenated conserved single-copy marker proteins were conducted to infer the taxonomic status of the current representatives of the class Halobacteria at the family and order levels. The current 76 genera with validly published names of the class Halobacteria were able to be assigned into eight families in one order. On the basis of these results, it is proposed that the current species with validly published names of the class Halobacteria should be remerged into the order Halobacteriales, then assigned to eight families, Haladaptataceae, Haloarculaceae, Halobacteriaceae, Halococcaceae, Haloferacaceae, Natronoarchaeaceae, Natrialbaceae and Halorubellaceae. Thus, Haladaptataceae fam. nov. is described based on Haladaptatus, Halomicrococcus and Halorussus and Halorubellaceae fam. nov. is proposed incorporating Haloarchaeobius and Halorubellus, respectively.

Multi-access edge computing (MEC) is regarded as a promising technology in the sixth-generation communication. However, the antenna gain is always affected by the environment when unmanned aerial vehicles (UAVs) are served as MEC platforms, resulting in unexpected channel errors. In order to deal with the problem and reduce the power consumption in the UAV-based MEC, we jointly optimize the access scheme and power allocation in the hierarchical UAV-based MEC. Specifically, UAVs are deployed in the lower layer to collect data from ground users. Moreover, a UAV with powerful computation ability is deployed in the upper layer to assist with computing. The goal is to guarantee the quality of service and minimize the total power consumption. We consider the errors caused by various perturbations in realistic circumstances and formulate a distributionally robust chance-constrained optimization problem with an uncertainty set. The problem with chance constraints is intractable. To tackle this issue, we utilize the conditional value-at-risk method to reformulate the problem into a semidefinite programming form. Then, a joint algorithm for access scheme and power allocation is designed. Finally, we conduct simulations to demonstrate the efficiency of the proposed algorithm.

Diffusion-weighted magnetic resonance imaging (DW-MRI) is a critical imaging method for capturing and modeling tissue microarchitecture at a millimeter scale. A common practice to model the measured DW-MRI signal is via fiber orientation distribution function (fODF). This function is the essential first step for the downstream tractography and connectivity analyses. With recent advantages in data sharing, large-scale multi-site DW-MRI datasets are being made available for multi-site studies. However, measurement variabilities (e.g., inter- and intra-site variability, hardware performance, and sequence design) are inevitable during the acquisition of DW-MRI. Most existing model-based methods (e.g., constrained spherical deconvolution (CSD)) and learning based methods (e.g., deep learning (DL)) do not explicitly consider such variabilities in fODF modeling, which consequently leads to inferior performance on multi-site and/or longitudinal diffusion studies. In this paper, we propose a novel data-driven deep constrained spherical deconvolution method to explicitly constrain the scan-rescan variabilities for a more reproducible and robust estimation of brain microstructure from repeated DW-MRI scans. Specifically, the proposed method introduces a new 3D volumetric scanner-invariant regularization scheme during the fODF estimation. We study the Human Connectome Project (HCP) young adults test-retest group as well as the MASiVar dataset (with inter- and intra-site scan/rescan data). The Baltimore Longitudinal Study of Aging (BLSA) dataset is employed for external validation. From the experimental results, the proposed data-driven framework outperforms the existing benchmarks in repeated fODF estimation. The proposed method is assessing the downstream connectivity analysis and shows increased performance in distinguishing subjects with different biomarkers., Comment: 33 pages, 7 figures

Multi-class cell segmentation in high-resolution Giga-pixel whole slide images (WSI) is critical for various clinical applications. Training such an AI model typically requires labor-intensive pixel-wise manual annotation from experienced domain experts (e.g., pathologists). Moreover, such annotation is error-prone when differentiating fine-grained cell types (e.g., podocyte and mesangial cells) via the naked human eye. In this study, we assess the feasibility of democratizing pathological AI deployment by only using lay annotators (annotators without medical domain knowledge). The contribution of this paper is threefold: (1) We proposed a molecular-empowered learning scheme for multi-class cell segmentation using partial labels from lay annotators; (2) The proposed method integrated Giga-pixel level molecular-morphology cross-modality registration, molecular-informed annotation, and molecular-oriented segmentation model, so as to achieve significantly superior performance via 3 lay annotators as compared with 2 experienced pathologists; (3) A deep corrective learning (learning with imperfect label) method is proposed to further improve the segmentation performance using partially annotated noisy data. From the experimental results, our learning method achieved F1 = 0.8496 using molecular-informed annotations from lay annotators, which is better than conventional morphology-based annotations (F1 = 0.7015) from experienced pathologists. Our method democratizes the development of a pathological segmentation deep model to the lay annotator level, which consequently scales up the learning process similar to a non-medical computer vision task. The official implementation and cell annotations are publicly available at https://github.com/hrlblab/MolecularEL.

Sensor fusion is a crucial augmentation technique for improving the accuracy and reliability of perception systems for automated vehicles under diverse driving conditions. However, adverse weather and low-light conditions remain challenging, where sensor performance degrades significantly, exposing vehicle safety to potential risks. Advanced sensors such as LiDARs can help mitigate the issue but with extremely high marginal costs. In this paper, we propose a novel transformer-based 3D object detection model "REDFormer" to tackle low visibility conditions, exploiting the power of a more practical and cost-effective solution by leveraging bird's-eye-view camera-radar fusion. Using the nuScenes dataset with multi-radar point clouds, weather information, and time-of-day data, our model outperforms state-of-the-art (SOTA) models on classification and detection accuracy. Finally, we provide extensive ablation studies of each model component on their contributions to address the above-mentioned challenges. Particularly, it is shown in the experiments that our model achieves a significant performance improvement over the baseline model in low-visibility scenarios, specifically exhibiting a 31.31% increase in rainy scenes and a 46.99% enhancement in nighttime scenes.The source code of this study is publicly available.

Deep learning has made great strides in medical imaging, enabled by hardware advances in GPUs. One major constraint for the development of new models has been the saturation of GPU memory resources during training. This is especially true in computational pathology, where images regularly contain more than 1 billion pixels. These pathological images are traditionally divided into small patches to enable deep learning due to hardware limitations. In this work, we explore whether the shared GPU/CPU memory architecture on the M1 Ultra systems-on-a-chip (SoCs) recently released by Apple, Inc. may provide a solution. These affordable systems (less than \$5000) provide access to 128 GB of unified memory (Mac Studio with M1 Ultra SoC). As a proof of concept for gigapixel deep learning, we identified tissue from background on gigapixel areas from whole slide images (WSIs). The model was a modified U-Net (4492 parameters) leveraging large kernels and high stride. The M1 Ultra SoC was able to train the model directly on gigapixel images (16000$\times$64000 pixels, 1.024 billion pixels) with a batch size of 1 using over 100 GB of unified memory for the process at an average speed of 1 minute and 21 seconds per batch with Tensorflow 2/Keras. As expected, the model converged with a high Dice score of 0.989 $\pm$ 0.005. Training up until this point took 111 hours and 24 minutes over 4940 steps. Other high RAM GPUs like the NVIDIA A100 (largest commercially accessible at 80 GB, $\sim$\$15000) are not yet widely available (in preview for select regions on Amazon Web Services at \$40.96/hour as a group of 8). This study is a promising step towards WSI-wise end-to-end deep learning with prevalent network architectures.

The segment anything model (SAM) was released as a foundation model for image segmentation. The promptable segmentation model was trained by over 1 billion masks on 11M licensed and privacy-respecting images. The model supports zero-shot image segmentation with various segmentation prompts (e.g., points, boxes, masks). It makes the SAM attractive for medical image analysis, especially for digital pathology where the training data are rare. In this study, we evaluate the zero-shot segmentation performance of SAM model on representative segmentation tasks on whole slide imaging (WSI), including (1) tumor segmentation, (2) non-tumor tissue segmentation, (3) cell nuclei segmentation. Core Results: The results suggest that the zero-shot SAM model achieves remarkable segmentation performance for large connected objects. However, it does not consistently achieve satisfying performance for dense instance object segmentation, even with 20 prompts (clicks/boxes) on each image. We also summarized the identified limitations for digital pathology: (1) image resolution, (2) multiple scales, (3) prompt selection, and (4) model fine-tuning. In the future, the few-shot fine-tuning with images from downstream pathological segmentation tasks might help the model to achieve better performance in dense object segmentation.

Analyzing high resolution whole slide images (WSIs) with regard to information across multiple scales poses a significant challenge in digital pathology. Multi-instance learning (MIL) is a common solution for working with high resolution images by classifying bags of objects (i.e. sets of smaller image patches). However, such processing is typically performed at a single scale (e.g., 20x magnification) of WSIs, disregarding the vital inter-scale information that is key to diagnoses by human pathologists. In this study, we propose a novel cross-scale MIL algorithm to explicitly aggregate inter-scale relationships into a single MIL network for pathological image diagnosis. The contribution of this paper is three-fold: (1) A novel cross-scale MIL (CS-MIL) algorithm that integrates the multi-scale information and the inter-scale relationships is proposed; (2) A toy dataset with scale-specific morphological features is created and released to examine and visualize differential cross-scale attention; (3) Superior performance on both in-house and public datasets is demonstrated by our simple cross-scale MIL strategy. The official implementation is publicly available at https://github.com/hrlblab/CS-MIL.

Semi-supervised domain adaptation (SSDA) adapts a learner to a new domain by effectively utilizing source domain data and a few labeled target samples. It is a practical yet under-investigated research topic. In this paper, we analyze the SSDA problem from two perspectives that have previously been overlooked, and correspondingly decompose it into two \emph{key subproblems}: \emph{robust domain adaptation (DA) learning} and \emph{maximal cross-domain data utilization}. \textbf{(i)} From a causal theoretical view, a robust DA model should distinguish the invariant ``concept'' (key clue to image label) from the nuisance of confounding factors across domains. To achieve this goal, we propose to generate \emph{concept-invariant samples} to enable the model to classify the samples through causal intervention, yielding improved generalization guarantees; \textbf{(ii)} Based on the robust DA theory, we aim to exploit the maximal utilization of rich source domain data and a few labeled target samples to boost SSDA further. Consequently, we propose a collaboratively debiasing learning framework that utilizes two complementary semi-supervised learning (SSL) classifiers to mutually exchange their unbiased knowledge, which helps unleash the potential of source and target domain training data, thereby producing more convincing pseudo-labels. Such obtained labels facilitate cross-domain feature alignment and duly improve the invariant concept learning. In our experimental study, we show that the proposed model significantly outperforms SOTA methods in terms of effectiveness and generalisability on SSDA datasets.

The vertical shear instability (VSI) is a promising mechanism to drive turbulence in protoplanetary disks. Numerical simulations in the literature demonstrate that the VSI non-linear saturation is predominated by the linear corrugation modes. These modes possess vertical wavelengths crucially longer than radial wavelengths. This paper aims to investigate the natural radial wavelength of corrugation modes upon VSI saturation, by a series of numerical simulations conducted in Athena++ at different grid resolutions, disk aspect ratios, and viscosity parameterized by $\nu$. We find a sign of convergence emerges at 64 cells per gas scale height for fiducial simulations, below which a continuous reduction of wavelengths with grid resolution is observed. Synthetic ALMA molecular line observations of $^{12}\rm CO(2-1)$ are performed to inspect the observability of the corrugation modes feature, which is significantly diminished with more than 32 cells per scale height. Flared and viscous disks, exhibiting longer saturation wavelengths, may mitigate the observational difficulty., Comment: 9 pages, 5 figures, accepted for publication in MNRAS

Recent studies have shown that the large-scale gas dynamics of protoplanetary disks (PPDs) are controlled by non-ideal magneto-hydrodynamics (MHD), but how this influences dust dynamics is not fully understood. To this end, we investigate the stability of dusty, magnetized disks subject to the Hall effect, which applies to planet-forming regions of PPDs. We find a novel Background Drift Hall Instability (BDHI) that may facilitate planetesimal formation in Hall-effected disk regions. Through a combination of linear analysis and nonlinear simulations, we demonstrate the viability and characteristics of BDHI. We find it can potentially dominate over the classical streaming instability (SI) and standard MHD instabilities at low dust-to-gas ratios and weak magnetic fields. We also identify magnetized versions of the classic SI, but these are usually subdominant. We highlight the complex interplay between magnetic fields and dust-gas dynamics in PPDs, underscoring the need to consider non-ideal MHD like the Hall effect in the broader narrative of planet formation., Comment: 23 pages, 12 figures, accepted by ApJ. Welcome any comments and suggestions!

Integrating large language models (LLMs) in autonomous vehicles enables conversation with AI systems to drive the vehicle. However, it also emphasizes the requirement for such systems to comprehend commands accurately and achieve higher-level personalization to adapt to the preferences of drivers or passengers over a more extended period. In this paper, we introduce an LLM-based framework, Talk2Drive, capable of translating natural verbal commands into executable controls and learning to satisfy personal preferences for safety, efficiency, and comfort with a proposed memory module. This is the first-of-its-kind multi-scenario field experiment that deploys LLMs on a real-world autonomous vehicle. Experiments showcase that the proposed system can comprehend human intentions at different intuition levels, ranging from direct commands like "can you drive faster" to indirect commands like "I am really in a hurry now". Additionally, we use the takeover rate to quantify the trust of human drivers in the LLM-based autonomous driving system, where Talk2Drive significantly reduces the takeover rate in highway, intersection, and parking scenarios. We also validate that the proposed memory module considers personalized preferences and further reduces the takeover rate by up to 65.2% compared with those without a memory module. The experiment video can be watched at https://www.youtube.com/watch?v=4BWsfPaq1Ro

Autonomous driving (AD) has made significant strides in recent years. However, existing frameworks struggle to interpret and execute spontaneous user instructions, such as "overtake the car ahead." Large Language Models (LLMs) have demonstrated impressive reasoning capabilities showing potential to bridge this gap. In this paper, we present LaMPilot, a novel framework that integrates LLMs into AD systems, enabling them to follow user instructions by generating code that leverages established functional primitives. We also introduce LaMPilot-Bench, the first benchmark dataset specifically designed to quantitatively evaluate the efficacy of language model programs in AD. Adopting the LaMPilot framework, we conduct extensive experiments to assess the performance of off-the-shelf LLMs on LaMPilot-Bench. Our results demonstrate the potential of LLMs in handling diverse driving scenarios and following user instructions in driving. To facilitate further research in this area, we release our code and data at https://github.com/PurdueDigitalTwin/LaMPilot., Comment: CVPR 2024

Joint rich and normalized automatic speech recognition (ASR), that produces transcriptions both with and without punctuation and capitalization, remains a challenge. End-to-end (E2E) ASR models offer both convenience and the ability to perform such joint transcription of speech. Training such models requires paired speech and rich text data, which is not widely available. In this paper, we compare two different approaches to train a stateless Transducer-based E2E joint rich and normalized ASR system, ready for streaming applications, with a limited amount of rich labeled data. The first approach uses a language model to generate pseudo-rich transcriptions of normalized training data. The second approach uses a single decoder conditioned on the type of the output. The first approach leads to E2E rich ASR which perform better on out-of-domain data, with up to 9% relative reduction in errors. The second approach demonstrates the feasibility of an E2E joint rich and normalized ASR system using as low as 5% rich training data with moderate (2.42% absolute) increase in errors., Comment: Submitted to ICASSP 2024

With the emergence of Large Language Models (LLMs) and Vision Foundation Models (VFMs), multimodal AI systems benefiting from large models have the potential to equally perceive the real world, make decisions, and control tools as humans. In recent months, LLMs have shown widespread attention in autonomous driving and map systems. Despite its immense potential, there is still a lack of a comprehensive understanding of key challenges, opportunities, and future endeavors to apply in LLM driving systems. In this paper, we present a systematic investigation in this field. We first introduce the background of Multimodal Large Language Models (MLLMs), the multimodal models development using LLMs, and the history of autonomous driving. Then, we overview existing MLLM tools for driving, transportation, and map systems together with existing datasets and benchmarks. Moreover, we summarized the works in The 1st WACV Workshop on Large Language and Vision Models for Autonomous Driving (LLVM-AD), which is the first workshop of its kind regarding LLMs in autonomous driving. To further promote the development of this field, we also discuss several important problems regarding using MLLMs in autonomous driving systems that need to be solved by both academia and industry.

Vehicle-to-vehicle (V2V) communications have greatly enhanced the perception capabilities of connected and automated vehicles (CAVs) by enabling information sharing to "see through the occlusions", resulting in significant performance improvements. However, developing and training complex multi-agent perception models from scratch can be expensive and unnecessary when existing single-agent models show remarkable generalization capabilities. In this paper, we propose a new framework termed MACP, which equips a single-agent pre-trained model with cooperation capabilities. We approach this objective by identifying the key challenges of shifting from single-agent to cooperative settings, adapting the model by freezing most of its parameters and adding a few lightweight modules. We demonstrate in our experiments that the proposed framework can effectively utilize cooperative observations and outperform other state-of-the-art approaches in both simulated and real-world cooperative perception benchmarks while requiring substantially fewer tunable parameters with reduced communication costs. Our source code is available at https://github.com/PurdueDigitalTwin/MACP., Comment: Accepted by WACV 2024, 10 pages, 8 figures, 4 tables

We present an end-to-end multichannel speaker-attributed automatic speech recognition (MC-SA-ASR) system that combines a Conformer-based encoder with multi-frame crosschannel attention and a speaker-attributed Transformer-based decoder. To the best of our knowledge, this is the first model that efficiently integrates ASR and speaker identification modules in a multichannel setting. On simulated mixtures of LibriSpeech data, our system reduces the word error rate (WER) by up to 12% and 16% relative compared to previously proposed single-channel and multichannel approaches, respectively. Furthermore, we investigate the impact of different input features, including multichannel magnitude and phase information, on the ASR performance. Finally, our experiments on the AMI corpus confirm the effectiveness of our system for real-world multichannel meeting transcription., Comment: 2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU 2023), Dec 2023, Taipei, Taiwan

The fusion of human-centric design and artificial intelligence (AI) capabilities has opened up new possibilities for next-generation autonomous vehicles that go beyond transportation. These vehicles can dynamically interact with passengers and adapt to their preferences. This paper proposes a novel framework that leverages Large Language Models (LLMs) to enhance the decision-making process in autonomous vehicles. By utilizing LLMs' linguistic and contextual understanding abilities with specialized tools, we aim to integrate the language and reasoning capabilities of LLMs into autonomous vehicles. Our research includes experiments in HighwayEnv, a collection of environments for autonomous driving and tactical decision-making tasks, to explore LLMs' interpretation, interaction, and reasoning in various scenarios. We also examine real-time personalization, demonstrating how LLMs can influence driving behaviors based on verbal commands. Our empirical results highlight the substantial advantages of utilizing chain-of-thought prompting, leading to improved driving decisions, and showing the potential for LLMs to enhance personalized driving experiences through ongoing verbal feedback. The proposed framework aims to transform autonomous vehicle operations, offering personalized support, transparent decision-making, and continuous learning to enhance safety and effectiveness. We achieve user-centric, transparent, and adaptive autonomous driving ecosystems supported by the integration of LLMs into autonomous vehicles., Comment: arXiv admin note: text overlap with arXiv:2309.10228

The future of autonomous vehicles lies in the convergence of human-centric design and advanced AI capabilities. Autonomous vehicles of the future will not only transport passengers but also interact and adapt to their desires, making the journey comfortable, efficient, and pleasant. In this paper, we present a novel framework that leverages Large Language Models (LLMs) to enhance autonomous vehicles' decision-making processes. By integrating LLMs' natural language capabilities and contextual understanding, specialized tools usage, synergizing reasoning, and acting with various modules on autonomous vehicles, this framework aims to seamlessly integrate the advanced language and reasoning capabilities of LLMs into autonomous vehicles. The proposed framework holds the potential to revolutionize the way autonomous vehicles operate, offering personalized assistance, continuous learning, and transparent decision-making, ultimately contributing to safer and more efficient autonomous driving technologies.

Crohn's disease (CD) is a chronic and relapsing inflammatory condition that affects segments of the gastrointestinal tract. CD activity is determined by histological findings, particularly the density of neutrophils observed on Hematoxylin and Eosin stains (H&E) imaging. However, understanding the broader morphometry and local cell arrangement beyond cell counting and tissue morphology remains challenging. To address this, we characterize six distinct cell types from H&E images and develop a novel approach for the local spatial signature of each cell. Specifically, we create a 10-cell neighborhood matrix, representing neighboring cell arrangements for each individual cell. Utilizing t-SNE for non-linear spatial projection in scatter-plot and Kernel Density Estimation contour-plot formats, our study examines patterns of differences in the cellular environment associated with the odds ratio of spatial patterns between active CD and control groups. This analysis is based on data collected at the two research institutes. The findings reveal heterogeneous nearest-neighbor patterns, signifying distinct tendencies of cell clustering, with a particular focus on the rectum region. These variations underscore the impact of data heterogeneity on cell spatial arrangements in CD patients. Moreover, the spatial distribution disparities between the two research sites highlight the significance of collaborative efforts among healthcare organizations. All research analysis pipeline tools are available at https://github.com/MASILab/cellNN., Comment: Submitted to SPIE Medical Imaging. San Diego, CA. February 2024

Podocytes, specialized epithelial cells that envelop the glomerular capillaries, play a pivotal role in maintaining renal health. The current description and quantification of features on pathology slides are limited, prompting the need for innovative solutions to comprehensively assess diverse phenotypic attributes within Whole Slide Images (WSIs). In particular, understanding the morphological characteristics of podocytes, terminally differentiated glomerular epithelial cells, is crucial for studying glomerular injury. This paper introduces the Spatial Pathomics Toolkit (SPT) and applies it to podocyte pathomics. The SPT consists of three main components: (1) instance object segmentation, enabling precise identification of podocyte nuclei; (2) pathomics feature generation, extracting a comprehensive array of quantitative features from the identified nuclei; and (3) robust statistical analyses, facilitating a comprehensive exploration of spatial relationships between morphological and spatial transcriptomics features.The SPT successfully extracted and analyzed morphological and textural features from podocyte nuclei, revealing a multitude of podocyte morphomic features through statistical analysis. Additionally, we demonstrated the SPT's ability to unravel spatial information inherent to podocyte distribution, shedding light on spatial patterns associated with glomerular injury. By disseminating the SPT, our goal is to provide the research community with a powerful and user-friendly resource that advances cellular spatial pathomics in renal pathology. The implementation and its complete source code of the toolkit are made openly accessible at https://github.com/hrlblab/spatial_pathomics.

Many anomaly detection approaches, especially deep learning methods, have been recently developed to identify abnormal image morphology by only employing normal images during training. Unfortunately, many prior anomaly detection methods were optimized for a specific "known" abnormality (e.g., brain tumor, bone fraction, cell types). Moreover, even though only the normal images were used in the training process, the abnormal images were often employed during the validation process (e.g., epoch selection, hyper-parameter tuning), which might leak the supposed ``unknown" abnormality unintentionally. In this study, we investigated these two essential aspects regarding universal anomaly detection in medical images by (1) comparing various anomaly detection methods across four medical datasets, (2) investigating the inevitable but often neglected issues on how to unbiasedly select the optimal anomaly detection model during the validation phase using only normal images, and (3) proposing a simple decision-level ensemble method to leverage the advantage of different kinds of anomaly detection without knowing the abnormality. The results of our experiments indicate that none of the evaluated methods consistently achieved the best performance across all datasets. Our proposed method enhanced the robustness of performance in general (average AUC 0.956).

The Segment Anything Model (SAM) is a recently proposed prompt-based segmentation model in a generic zero-shot segmentation approach. With the zero-shot segmentation capacity, SAM achieved impressive flexibility and precision on various segmentation tasks. However, the current pipeline requires manual prompts during the inference stage, which is still resource intensive for biomedical image segmentation. In this paper, instead of using prompts during the inference stage, we introduce a pipeline that utilizes the SAM, called all-in-SAM, through the entire AI development workflow (from annotation generation to model finetuning) without requiring manual prompts during the inference stage. Specifically, SAM is first employed to generate pixel-level annotations from weak prompts (e.g., points, bounding box). Then, the pixel-level annotations are used to finetune the SAM segmentation model rather than training from scratch. Our experimental results reveal two key findings: 1) the proposed pipeline surpasses the state-of-the-art (SOTA) methods in a nuclei segmentation task on the public Monuseg dataset, and 2) the utilization of weak and few annotations for SAM finetuning achieves competitive performance compared to using strong pixel-wise annotated data.

Analysis of live-cell imaging experiments at the resolution of single cells provides exciting insights into the inner workings of biological systems. Advances in biological imaging and computer vision allow for segmentation of natural images with a high degree of accuracy. However, automation of the segmentation pipeline at the single cell resolution remains a challenging task. Complex deep learning models require large, well-annotated datasets that are rarely available in biology. In this research, we explore various methods that optimize state of the art deep learning frameworks, despite limited resources. We trained a large permutation of models to quantify their capacity and to measure the effects of temporal information, spatial awareness and transfer learning on model performance. We find that, although training set size is most impactful in improving model accuracy, we can leverage techniques like spatial awareness and transfer learning to compromise for the lack of data. These insights show that, with an abundance of data, light-weight models can be as performant as their heavy-weight counterparts in cellular analysis.

Analysis of live-cell imaging experiments at the resolution of single cells provides exciting insights into the inner workings of biological systems. Advances in biological imaging and computer vision allow for segmentation of natural images with a high degree of accuracy. However, automation of the segmentation pipeline at the single cell resolution remains a challenging task. Complex deep learning models require large, well-annotated datasets that are rarely available in biology. In this research, we explore various methods that optimize state of the art deep learning frameworks, despite limited resources. We trained a large permutation of models to quantify their capacity and to measure the effects of temporal information, spatial awareness and transfer learning on model performance. We find that, although training set size is most impactful in improving model accuracy, we can leverage techniques like spatial awareness and transfer learning to compromise for the lack of data. These insights show that, with an abundance of data, light-weight models can be as performant as their heavy-weight counterparts in cellular analysis.

Analysis of live-cell imaging experiments at the resolution of single cells provides exciting insights into the inner workings of biological systems. Advances in biological imaging and computer vision allow for segmentation of natural images with a high degree of accuracy. However, automation of the segmentation pipeline at the single cell resolution remains a challenging task. Complex deep learning models require large, well-annotated datasets that are rarely available in biology. In this research, we explore various methods that optimize state of the art deep learning frameworks, despite limited resources. We trained a large permutation of models to quantify their capacity and to measure the effects of temporal information, spatial awareness and transfer learning on model performance. We find that, although training set size is most impactful in improving model accuracy, we can leverage techniques like spatial awareness and transfer learning to compromise for the lack of data. These insights show that, with an abundance of data, light-weight models can be as performant as their heavy-weight counterparts in cellular analysis.

The vertical shear instability (VSI) is a robust and potentially important phenomenon in irradiated protoplanetary disks (PPDs), yet the mechanism by which it saturates remains poorly understood. Global simulations suggest that the non-linear evolution of the VSI is dominated by radially propagating inertial wavetrains (called `body modes'), but these are known to be susceptible to a parametric instability. In this paper, we propose that the global VSI saturates via this secondary instability, which initiates a redistribution of energy from the large scales to smaller-scale inertial waves, and finally into a turbulent cascade. We present an analytic theory of the instability in a simple idealised model that captures the main physical and mathematical details of the problem. In addition, we conduct numerical simulations with the SNOOPY code to consolidate the theory. Once the parametric instability prevails, the VSI is likely far more disordered and incoherent than current global simulations suggest. We also argue that it is challenging to capture parametric instability in global simulations unless the radial resolution is very fine, possibly $\sim 300$ grid cells per scale height in radius., Comment: 10 pages, 2 figures, submitted to MNRAS

The vertical shear instability (VSI) is a robust phenomenon in irradiated protoplanetary discs (PPDs). The majority of previous numerical simulations have focused on the turbulent properties of its saturated state. However, the saturation of the VSI manifests as large-scale coherent radially travelling inertial waves. In this paper, we study inertial-wave-disc interactions and their impact on VSI saturation. Inertial-wave linear theory is developed and applied to a representative global 2D simulation using the ATHENA++ code. It is found that the VSI saturates by separating the disc into several radial wave zones roughly demarcated by Lindblad resonances (turning points); this structure also manifests in a modest radial variation in the vertical turbulence strength. Future numerical work should employ large radial domains to accommodate this radial structure of the VSI, while concurrently adopting sufficiently fine resolutions to resolve the parametric instability that attacks the saturated VSI inertial waves.

Multi-instance learning (MIL) is widely used in the computer-aided interpretation of pathological Whole Slide Images (WSIs) to solve the lack of pixel-wise or patch-wise annotations. Often, this approach directly applies "natural image driven" MIL algorithms which overlook the multi-scale (i.e. pyramidal) nature of WSIs. Off-the-shelf MIL algorithms are typically deployed on a single-scale of WSIs (e.g., 20x magnification), while human pathologists usually aggregate the global and local patterns in a multi-scale manner (e.g., by zooming in and out between different magnifications). In this study, we propose a novel cross-scale attention mechanism to explicitly aggregate inter-scale interactions into a single MIL network for Crohn's Disease (CD), which is a form of inflammatory bowel disease. The contribution of this paper is two-fold: (1) a cross-scale attention mechanism is proposed to aggregate features from different resolutions with multi-scale interaction; and (2) differential multi-scale attention visualizations are generated to localize explainable lesion patterns. By training ~250,000 H&E-stained Ascending Colon (AC) patches from 20 CD patient and 30 healthy control samples at different scales, our approach achieved a superior Area under the Curve (AUC) score of 0.8924 compared with baseline models. The official implementation is publicly available at https://github.com/hrlblab/CS-MIL.

The rapid development of diagnostic technologies in healthcare is leading to higher requirements for physicians to handle and integrate the heterogeneous, yet complementary data that are produced during routine practice. For instance, the personalized diagnosis and treatment planning for a single cancer patient relies on the various images (e.g., radiological, pathological, and camera images) and non-image data (e.g., clinical data and genomic data). However, such decision-making procedures can be subjective, qualitative, and have large inter-subject variabilities. With the recent advances in multi-modal deep learning technologies, an increasingly large number of efforts have been devoted to a key question: how do we extract and aggregate multi-modal information to ultimately provide more objective, quantitative computer-aided clinical decision making? This paper reviews the recent studies on dealing with such a question. Briefly, this review will include the (1) overview of current multi-modal learning workflows, (2) summarization of multi-modal fusion methods, (3) discussion of the performance, (4) applications in disease diagnosis and prognosis, and (5) challenges and future directions.

Multiple Sclerosis (MS) is a chronic neuroinflammatory disease and multi-modality MRIs are routinely used to monitor MS lesions. Many automatic MS lesion segmentation models have been developed and have reached human-level performance. However, most established methods assume the MRI modalities used during training are also available during testing, which is not guaranteed in clinical practice. Previously, a training strategy termed Modality Dropout (ModDrop) has been applied to MS lesion segmentation to achieve the state-of-the-art performance with missing modality. In this paper, we present a novel method dubbed ModDrop++ to train a unified network adaptive to an arbitrary number of input MRI sequences. ModDrop++ upgrades the main idea of ModDrop in two key ways. First, we devise a plug-and-play dynamic head and adopt a filter scaling strategy to improve the expressiveness of the network. Second, we design a co-training strategy to leverage the intra-subject relation between full modality and missing modality. Specifically, the intra-subject co-training strategy aims to guide the dynamic head to generate similar feature representations between the full- and missing-modality data from the same subject. We use two public MS datasets to show the superiority of ModDrop++. Source code and trained models are available at https://github.com/han-liu/ModDropPlusPlus., Comment: MICCAI 2022

Integrating cross-department multi-modal data (e.g., radiological, pathological, genomic, and clinical data) is ubiquitous in brain cancer diagnosis and survival prediction. To date, such an integration is typically conducted by human physicians (and panels of experts), which can be subjective and semi-quantitative. Recent advances in multi-modal deep learning, however, have opened a door to leverage such a process to a more objective and quantitative manner. Unfortunately, the prior arts of using four modalities on brain cancer survival prediction are limited by a "complete modalities" setting (i.e., with all modalities available). Thus, there are still open questions on how to effectively predict brain cancer survival from the incomplete radiological, pathological, genomic, and demographic data (e.g., one or more modalities might not be collected for a patient). For instance, should we use both complete and incomplete data, and more importantly, how to use those data? To answer the preceding questions, we generalize the multi-modal learning on cross-department multi-modal data to a missing data setting. Our contribution is three-fold: 1) We introduce optimal multi-modal learning with missing data (MMD) pipeline with optimized hardware consumption and computational efficiency; 2) We extend multi-modal learning on radiological, pathological, genomic, and demographic data into missing data scenarios; 3) a large-scale public dataset (with 962 patients) is collected to systematically evaluate glioma tumor survival prediction using four modalities. The proposed method improved the C-index of survival prediction from 0.7624 to 0.8053.

Automatic methods to segment the vestibular schwannoma (VS) tumors and the cochlea from magnetic resonance imaging (MRI) are critical to VS treatment planning. Although supervised methods have achieved satisfactory performance in VS segmentation, they require full annotations by experts, which is laborious and time-consuming. In this work, we aim to tackle the VS and cochlea segmentation problem in an unsupervised domain adaptation setting. Our proposed method leverages both the image-level domain alignment to minimize the domain divergence and semi-supervised training to further boost the performance. Furthermore, we propose to fuse the labels predicted from multiple models via noisy label correction. In the MICCAI 2021 crossMoDA challenge, our results on the final evaluation leaderboard showed that our proposed method has achieved promising segmentation performance with mean dice score of 79.9% and 82.5% and ASSD of 1.29 mm and 0.18 mm for VS tumor and cochlea, respectively. The cochlea ASSD achieved by our method has outperformed all other competing methods as well as the supervised nnU-Net., Comment: Accepted by MICCAI 2021 BrainLes Workshop. arXiv admin note: substantial text overlap with arXiv:2109.06274

The outer protoplanetary disks (PPDs) can be subject to the magnetorotational instability (MRI) and the vertical shear instability (VSI). While both processes can drive turbulence in the disk, existing numerical simulations have studied them separately. In this paper, we conduct global 3D non-ideal magnetohydrodynamic (MHD) simulations for outer PPDs with ambipolar diffusion and instantaneous cooling, and hence conductive to both instabilities. Given the range of ambipolar Els\"{a}sser numbers ($Am$) explored, it is found that the VSI turbulence dominates over the MRI when ambipolar diffusion is strong ($Am=0.1$); the VSI and MRI can co-exist for $Am=1$; and the VSI is overwhelmed by the MRI when ambipolar diffusion is weak ($Am=10$). Angular momentum transport process is primarily driven by MHD winds, while viscous accretion due to MRI and/or VSI turbulence makes a moderate contribution in most cases. Spontaneous magnetic flux concentration and formation of annular substructures remain robust in strong ambipolar diffusion dominated disks ($Am\leq1$) with the presence of the VSI. Ambipolar diffusion is the major contributor to the magnetic flux concentration phenomenon rather than advection., Comment: 9 pages, 10 figures, accepted for publication in MNRAS

Turbulent motions are believed to regulate angular momentum transport and influence dust evolution in protoplanetary disks. Measuring the strength of turbulence is challenging through gas line observations because of the requirement for high spatial and spectral resolution data, and an exquisite determination of the temperature. In this work, taking the well-known HD 163296 disk as an example, we investigated the contrast of gaps identified in high angular resolution continuum images as a probe for the level of turbulence. With self-consistent radiative transfer models, we simultaneously analyzed the radial brightness profiles along the disk major and minor axes, and the azimuthal brightness profiles of the B67 and B100 rings. By fitting all the gap contrasts measured from these profiles, we constrained the gas-to-dust scale height ratio $\Lambda$ to be $3.0_{-0.8}^{+0.3}$, $1.2_{-0.1}^{+0.1}$ and ${\ge}\,6.5$ for the D48, B67 and B100 regions, respectively. The varying gas-to-dust scale height ratios indicate that the degree of dust settling changes with radius. The inferred values for $\Lambda$ translate into a turbulence level of $\alpha_{\rm turb}\,{<}\,3\times10^{-3}$ in the D48 and B100 regions, which is consistent with previous upper limits set by gas line observations. However, turbulent motions in the B67 ring are strong with $\alpha_{\rm turb}\,{\sim}1.2\,{\times}\,10^{-2}$. Due to the degeneracy between $\Lambda$ and the depth of dust surface density drops, the turbulence strength in the D86 gap region is not constrained., Comment: 13 pages, 7 figures, accepted for publication in Science China Physics, Mechanics & Astronomy

© 2020, The Author(s), under exclusive licence to Springer Nature America, Inc. Inducible expression of neoantigens in mice would enable the study of endogenous antigen-specific naïve T cell responses in disease and infection, but has been difficult to generate because leaky antigen expression in the thymus results in central T cell tolerance. Here we develop inversion-induced joined neoantigen (NINJA), using RNA splicing, DNA recombination and three levels of regulation to prevent leakiness and allow tight control over neoantigen expression. We apply NINJA to create tumor cell lines with inducible neoantigen expression, which could be used to study antitumor immunity. We also show that the genetic regulation in NINJA mice bypasses central and peripheral tolerance mechanisms and allows for robust endogenous CD8 and CD4 T cell responses on neoantigen induction in peripheral tissues. NINJA will enable studies of how T cells respond to defined neoantigens in the context of peripheral tolerance, transplantation, autoimmune diseases and cancer.

The vertical shear instability (VSI) is a robust phenomenon in irradiated protoplanetary disks (PPDs). The majority of previous numerical simulations have focused on the turbulent properties of its saturated state. However, the saturation of the VSI manifests as large-scale coherent radially travelling inertial waves. In this paper, we study inertial-wave-disk interactions and their impact on VSI saturation. Inertial-wave linear theory is developed and applied to a representative global 2D simulation using the Athena++ code. It is found that the VSI saturates by separating the disk into several radial wave zones roughly demarcated by corotation resonances (turning points); this structure also manifests in modest radial variations in the vertical turbulence strength. Future numerical work should employ large radial domains to accommodate this radial structure of the VSI, while concurrently adopting sufficiently fine resolutions to resolve the parametric instability that attacks the saturated VSI inertial waves., Comment: 8 pages, 6 figures, accepted for publication in MNRAS

The vertical shear instability (VSI) is a robust and potentially important phenomenon in irradiated protoplanetary disks (PPDs), yet the mechanism by which it saturates remains poorly understood. Global simulations suggest that the non-linear evolution of the VSI is dominated by radially propagating inertial wavetrains (called `body modes'), but these are known to be susceptible to a parametric instability. In this paper, we propose that the global VSI saturates via this secondary instability, which initiates a redistribution of energy from the large scales to smaller-scale inertial waves, and finally into a turbulent cascade. We present an analytic theory of the instability in a simple idealised model that captures the main physical and mathematical details of the problem. In addition, we conduct numerical simulations with the SNOOPY code to consolidate the theory. Once the parametric instability prevails, the VSI is likely far more disordered and incoherent than current global simulations suggest. We also argue that it is challenging to capture parametric instability in global simulations unless the radial resolution is very fine, possibly $\sim 300$ grid cells per scale height in radius., Comment: 10 pages, 2 figures, submitted to MNRAS

Comprehensive semantic segmentation on renal pathological images is challenging due to the heterogeneous scales of the objects. For example, on a whole slide image (WSI), the cross-sectional areas of glomeruli can be 64 times larger than that of the peritubular capillaries, making it impractical to segment both objects on the same patch, at the same scale. To handle this scaling issue, prior studies have typically trained multiple segmentation networks in order to match the optimal pixel resolution of heterogeneous tissue types. This multi-network solution is resource-intensive and fails to model the spatial relationship between tissue types. In this paper, we propose the Omni-Seg+ network, a scale-aware dynamic neural network that achieves multi-object (six tissue types) and multi-scale (5X to 40X scale) pathological image segmentation via a single neural network. The contribution of this paper is three-fold: (1) a novel scale-aware controller is proposed to generalize the dynamic neural network from single-scale to multi-scale; (2) semi-supervised consistency regularization of pseudo-labels is introduced to model the inter-scale correlation of unannotated tissue types into a single end-to-end learning paradigm; and (3) superior scale-aware generalization is evidenced by directly applying a model trained on human kidney images to mouse kidney images, without retraining. By learning from ~150,000 human pathological image patches from six tissue types at three different resolutions, our approach achieved superior segmentation performance according to human visual assessment and evaluation of image-omics (i.e., spatial transcriptomics). The official implementation is available at https://github.com/ddrrnn123/Omni-Seg.