Correctly assessing the malignancy of breast lesions identified during ultrasound examinations is crucial for effective clinical decision-making. However, the current "golden standard" relies on manual BI-RADS scoring by clinicians, often leading to unnecessary biopsies and a significant mental health burden on patients and their families. In this paper, we introduce PersonalizedUS, an interpretable machine learning system that leverages recent advances in conformal prediction to provide precise and personalized risk estimates with local coverage guarantees and sensitivity, specificity, and predictive values above 0.9 across various threshold levels. In particular, we identify meaningful lesion subgroups where distribution-free, model-agnostic conditional coverage holds, with approximately 90% of our prediction sets containing only the ground truth in most lesion subgroups, thus explicitly characterizing for which patients the model is most suitably applied. Moreover, we make available a curated tabular dataset of 1936 biopsied breast lesions from a recent observational multicenter study and benchmark the performance of several state-of-the-art learning algorithms. We also report a successful case study of the deployed system in the same multicenter context. Concrete clinical benefits include up to a 65% reduction in requested biopsies among BI-RADS 4a and 4b lesions, with minimal to no missed cancer cases.
This research introduces an innovative method for the early screening of cardiorespiratory diseases based on an acquisition protocol, which leverages commodity smartphone's Inertial Measurement Units (IMUs) and deep learning techniques. We collected, in a clinical setting, a dataset featuring recordings of breathing kinematics obtained by accelerometer and gyroscope readings from five distinct body regions. We propose an end-to-end deep learning pipeline for early cardiorespiratory disease screening, incorporating a preprocessing step segmenting the data into individual breathing cycles, and a recurrent bidirectional module capturing features from diverse body regions. We employed Leave-one-out-cross-validation with Bayesian optimization for hyperparameter tuning and model selection. The experimental results consistently demonstrated the superior performance of a bidirectional Long-Short Term Memory (Bi-LSTM) as a feature encoder architecture, yielding an average sensitivity of $0.81 \pm 0.02$, specificity of $0.82 \pm 0.05$, F1 score of $0.81 \pm 0.02$, and accuracy of $80.2\% \pm 3.9$ across diverse seed variations. We also assessed generalization capabilities on a skewed distribution, comprising exclusively healthy patients not used in training, revealing a true negative rate of $74.8 \% \pm 4.5$. The sustained accuracy of predictions over time during breathing cycles within a single patient underscores the efficacy of the preprocessing strategy, highlighting the model's ability to discern significant patterns throughout distinct phases of the respiratory cycle. This investigation underscores the potential usefulness of widely available smartphones as devices for timely cardiorespiratory disease screening in the general population, in at-home settings, offering crucial assistance to public health efforts (especially during a pandemic outbreaks, such as the recent COVID-19).
District Heating (DH) systems are essential for energy-efficient urban heating. However, despite the advancements in automated fault detection and diagnosis (FDD), DH still faces challenges in operational faults that impact efficiency. This study introduces the Shared Nearest Neighbor Enhanced District Heating Anomaly Detection (SHEDAD) approach, designed to approximate the DH network topology and allow for local anomaly detection without disclosing sensitive information, such as substation locations. The approach leverages a multi-adaptive k-Nearest Neighbor (k-NN) graph to improve the initial neighborhood creation. Moreover, it introduces a merging technique that reduces noise and eliminates trivial edges. We use the Median Absolute Deviation (MAD) and modified z-scores to flag anomalous substations. The results reveal that SHEDAD outperforms traditional clustering methods, achieving significantly lower intra-cluster variance and distance. Additionally, SHEDAD effectively isolates and identifies two distinct categories of anomalies: supply temperatures and substation performance. We identified 30 anomalous substations and reached a sensitivity of approximately 65\% and specificity of approximately 97\%. By focusing on this subset of poor-performing substations in the network, SHEDAD enables more targeted and effective maintenance interventions, which can reduce energy usage while optimizing network performance.
The storage and recall of factual associations in auto-regressive transformer language models (LMs) have drawn a great deal of attention, inspiring knowledge editing by directly modifying the located model weights. Most editing works achieve knowledge editing under the guidance of existing interpretations of knowledge recall that mainly focus on subject knowledge. However, these interpretations are seriously flawed, neglecting relation information and leading to the over-generalizing problem for editing. In this work, we discover a novel relation-focused perspective to interpret the knowledge recall of transformer LMs during inference and apply it on knowledge editing to avoid over-generalizing. Experimental results on the dataset supplemented with a new R-Specificity criterion demonstrate that our editing approach significantly alleviates over-generalizing while remaining competitive on other criteria, breaking the domination of subject-focused editing for future research.
This study asks how self-supervised speech models represent suprasegmental categories like Mandarin lexical tone, English lexical stress, and English phrasal accents. Through a series of probing tasks, we make layer-wise comparisons of English and Mandarin 12 layer monolingual models. Our findings suggest that 1) English and Mandarin wav2vec 2.0 models learn contextual representations of abstract suprasegmental categories which are strongest in the middle third of the network. 2) Models are better at representing features that exist in the language of their training data, and this difference is driven by enriched context in transformer blocks, not local acoustic representation. 3) Fine-tuned wav2vec 2.0 improves performance in later layers compared to pre-trained models mainly for lexically contrastive features like tone and stress, 4) HuBERT and WavLM learn similar representations to wav2vec 2.0, differing mainly in later layer performance. Our results extend previous understanding of how models represent suprasegmentals and offer new insights into the language-specificity and contextual nature of these representations.
Histological Tertiary Lymphoid Structures (TLSs) are increasingly recognized for their correlation with the efficacy of immunotherapy in various solid tumors. Traditionally, the identification and characterization of TLSs rely on immunohistochemistry (IHC) staining techniques, utilizing markers such as CD20 for B cells. Despite the specificity of IHC, Hematoxylin-Eosin (H&E) staining offers a more accessible and cost-effective choice. Capitalizing on the prevalence of H&E staining slides, we introduce a novel Mask-Guided Adversarial Transfer Learning method designed for virtual pathological staining. This method adeptly captures the nuanced color variations across diverse tissue types under various staining conditions, such as nucleus, red blood cells, positive reaction regions, without explicit label information, and adeptly synthesizes realistic IHC-like virtual staining patches, even replicating the positive reaction. Further, we propose the Virtual IHC Pathology Analysis Network (VIPA-Net), an integrated framework encompassing a Mask-Guided Transfer Module and an H&E-Based Virtual Staining TLS Detection Module. VIPA-Net synergistically harnesses both H\&E staining slides and the synthesized virtual IHC patches to enhance the detection of TLSs within H&E Whole Slide Images (WSIs). We evaluate the network with a comprehensive dataset comprising 1019 annotated slides from The Cancer Genome Atlas (TCGA). Experimental results compellingly illustrate that the VIPA-Net substantially elevates TLS detection accuracy, effectively circumventing the need for actual CD20 staining across the public dataset.
Question difficulty estimation remains a multifaceted challenge in educational and assessment settings. Traditional approaches often focus on surface-level linguistic features or learner comprehension levels, neglecting the intricate interplay of factors contributing to question complexity. This paper presents a novel framework for domain-specific question difficulty estimation, leveraging a suite of NLP techniques and knowledge graph analysis. We introduce four key parameters: Topic Retrieval Cost, Topic Salience, Topic Coherence, and Topic Superficiality, each capturing a distinct facet of question complexity within a given subject domain. These parameters are operationalized through topic modelling, knowledge graph analysis, and information retrieval techniques. A model trained on these features demonstrates the efficacy of our approach in predicting question difficulty. By operationalizing these parameters, our framework offers a novel approach to question complexity estimation, paving the way for more effective question generation, assessment design, and adaptive learning systems across diverse academic disciplines.
The emergence of text-to-image synthesis (TIS) models has significantly influenced digital image creation by producing high-quality visuals from written descriptions. Yet these models heavily rely on the quality and specificity of textual prompts, posing a challenge for novice users who may not be familiar with TIS-model-preferred prompt writing. Existing solutions relieve this via automatic model-preferred prompt generation from user queries. However, this single-turn manner suffers from limited user-centricity in terms of result interpretability and user interactivity. To address these issues, we propose DialPrompt, a multi-turn dialogue-based TIS prompt generation model that emphasises user-centricity. DialPrompt is designed to follow a multi-turn guidance workflow, where in each round of dialogue the model queries user with their preferences on possible optimization dimensions before generating the final TIS prompt. To achieve this, we mined 15 essential dimensions for high-quality prompts from advanced users and curated a multi-turn dataset. Through training on this dataset, DialPrompt can improve interpretability by allowing users to understand the correlation between specific phrases and image attributes. Additionally, it enables greater user control and engagement in the prompt generation process, leading to more personalized and visually satisfying outputs. Experiments indicate that DialPrompt achieves a competitive result in the quality of synthesized images, outperforming existing prompt engineering approaches by 5.7%. Furthermore, in our user evaluation, DialPrompt outperforms existing approaches by 46.5% in user-centricity score and is rated 7.9/10 by 19 human reviewers.
Pulmonary Fibrosis (PF) is an immedicable respiratory condition distinguished by permanent fibrotic alterations in the pulmonary tissue for which there is no cure. Hence, it is crucial to diagnose PF swiftly and precisely. The existing research on deep learning-based pulmonary fibrosis detection methods has limitations, including dataset sample sizes and a lack of standardization in data preprocessing and evaluation metrics. This study presents a comparative analysis of four vision transformers regarding their efficacy in accurately detecting and classifying patients with Pulmonary Fibrosis and their ability to localize abnormalities within Images obtained from Computerized Tomography (CT) scans. The dataset consisted of 13,486 samples selected out of 24647 from the Pulmonary Fibrosis dataset, which included both PF-positive CT and normal images that underwent preprocessing. The preprocessed images were divided into three sets: the training set, which accounted for 80% of the total pictures; the validation set, which comprised 10%; and the test set, which also consisted of 10%. The vision transformer models, including ViT, MobileViT2, ViTMSN, and BEiT were subjected to training and validation procedures, during which hyperparameters like the learning rate and batch size were fine-tuned. The overall performance of the optimized architectures has been assessed using various performance metrics to showcase the consistent performance of the fine-tuned model. Regarding performance, ViT has shown superior performance in validation and testing accuracy and loss minimization, specifically for CT images when trained at a single epoch with a tuned learning rate of 0.0001. The results were as follows: validation accuracy of 99.85%, testing accuracy of 100%, training loss of 0.0075, and validation loss of 0.0047. The experimental evaluation of the independently collected data gives empirical evidence that the optimized Vision Transformer (ViT) architecture exhibited superior performance compared to all other optimized architectures. It achieved a flawless score of 1.0 in various standard performance metrics, including Sensitivity, Specificity, Accuracy, F1-score, Precision, Recall, Mathew Correlation Coefficient (MCC), Precision-Recall Area under the Curve (AUC PR), Receiver Operating Characteristic and Area Under the Curve (ROC-AUC). Therefore, the optimized Vision Transformer (ViT) functions as a reliable diagnostic tool for the automated categorization of individuals with pulmonary fibrosis (PF) using chest computed
Introduction: Electrical impedance spectroscopy (EIS) has recently developed as a novel diagnostic device for screening and evaluating cervical dysplasia, prostate cancer, breast cancer and basal cell carcinoma. The current study aimed to validate and evaluate bioimpedance as a diagnostic tool for tobacco-induced oral lesions. Methodology: The study comprised 50 OSCC and OPMD tissue specimens for in-vitro study and 320 subjects for in vivo study. Bioimpedance device prepared and calibrated. EIS measurements were done for the habit and control groups and were compared. Results: The impedance value in the control group was significantly higher compared to the OPMD and OSCC groups. Diagnosis based on BIS measurements has a sensitivity of 95.9% and a specificity of 86.7%. Conclusion: Bioimpedance device can help in decision-making for differentiating OPMD and OSCC cases and their management, especially in primary healthcare settings. Keywords: Impedance, Cancer, Diagnosis, Device, Community
Retrieval Augmented Generation (RAG) is a technique used to augment Large Language Models (LLMs) with contextually relevant, time-critical, or domain-specific information without altering the underlying model parameters. However, constructing RAG systems that can effectively synthesize information from large and diverse set of documents remains a significant challenge. We introduce a novel data-centric RAG workflow for LLMs, transforming the traditional retrieve-then-read system into a more advanced prepare-then-rewrite-then-retrieve-then-read framework, to achieve higher domain expert-level understanding of the knowledge base. Our methodology relies on generating metadata and synthetic Questions and Answers (QA) for each document, as well as introducing the new concept of Meta Knowledge Summary (MK Summary) for metadata-based clusters of documents. The proposed innovations enable personalized user-query augmentation and in-depth information retrieval across the knowledge base. Our research makes two significant contributions: using LLMs as evaluators and employing new comparative performance metrics, we demonstrate that (1) using augmented queries with synthetic question matching significantly outperforms traditional RAG pipelines that rely on document chunking (p < 0.01), and (2) meta knowledge-augmented queries additionally significantly improve retrieval precision and recall, as well as the final answers breadth, depth, relevancy, and specificity. Our methodology is cost-effective, costing less than $20 per 2000 research papers using Claude 3 Haiku, and can be adapted with any fine-tuning of either the language or embedding models to further enhance the performance of end-to-end RAG pipelines.
Transparency in automated systems could be afforded through the provision of intelligible explanations. While transparency is desirable, might it lead to catastrophic outcomes (such as anxiety), that could outweigh its benefits? It's quite unclear how the specificity of explanations (level of transparency) influences recipients, especially in autonomous driving (AD). In this work, we examined the effects of transparency mediated through varying levels of explanation specificity in AD. We first extended a data-driven explainer model by adding a rule-based option for explanation generation in AD, and then conducted a within-subject lab study with 39 participants in an immersive driving simulator to study the effect of the resulting explanations. Specifically, our investigation focused on: (1) how different types of explanations (specific vs. abstract) affect passengers' perceived safety, anxiety, and willingness to take control of the vehicle when the vehicle perception system makes erroneous predictions; and (2) the relationship between passengers' behavioural cues and their feelings during the autonomous drives. Our findings showed that passengers felt safer with specific explanations when the vehicle's perception system had minimal errors, while abstract explanations that hid perception errors led to lower feelings of safety. Anxiety levels increased when specific explanations revealed perception system errors (high transparency). We found no significant link between passengers' visual patterns and their anxiety levels. Our study suggests that passengers prefer clear and specific explanations (high transparency) when they originate from autonomous vehicles (AVs) with optimal perceptual accuracy.
Deep Learning (DL) methods have been used for electrocardiogram (ECG) processing in a wide variety of tasks, demonstrating good performance compared with traditional signal processing algorithms. These methods offer an efficient framework with a limited need for apriori data pre-processing and feature engineering. While several studies use this approach for ECG signal delineation, a significant gap persists between the expected and the actual outcome. Existing methods rely on a sample-to-sample classifier. However, the clinical expected outcome consists of a set of onset, offset, and peak for the different waves that compose each R-R interval. To align the actual with the expected output, it is necessary to incorporate post-processing algorithms. This counteracts two of the main advantages of DL models, since these algorithms are based on assumptions and slow down the method's performance. In this paper, we present Keypoint Estimation for Electrocardiogram Delineation (KEED), a novel DL model designed for keypoint estimation, which organically offers an output aligned with clinical expectations. By standing apart from the conventional sample-to-sample classifier, we achieve two benefits: (i) Eliminate the need for additional post-processing, and (ii) Establish a flexible framework that allows the adjustment of the threshold value considering the sensitivity-specificity tradeoff regarding the particular clinical requirements. The proposed method's performance is compared with state-of-the-art (SOTA) signal processing methods. Remarkably, KEED significantly outperforms despite being optimized with an extremely limited annotated data. In addition, KEED decreases the inference time by a factor ranging from 52x to 703x.
In recent years, there has been an effort to extend the classical notion of phylogenetic balance, originally defined in the context of trees, to networks. One of the most natural ways to do this is with the so-called $B_2$ index. In this paper, we study the $B_2$ index for a prominent class of phylogenetic networks: galled trees. We show that the $B_2$ index of a uniform leaf-labeled galled tree converges in distribution as the network becomes large. We characterize the corresponding limiting distribution, and show that its expected value is 2.707911858984... This is the first time that a balance index has been studied to this level of detail for a random phylogenetic network. One specificity of this work is that we use two different and independent approaches, each with its advantages: analytic combinatorics, and local limits. The analytic combinatorics approach is more direct, as it relies on standard tools; but it involves slightly more complex calculations. Because it has not previously been used to study such questions, the local limit approach requires developing an extensive framework beforehand; however, this framework is interesting in itself and can be used to tackle other similar problems.
Atherosclerosis, a chronic inflammatory cardiovascular disease, leads to arterial constriction caused by the accumulation of lipids, cholesterol, and various substances within artery walls. Such plaque can rupture, resulting in a blood clot that obstructs major arteries and may initiate myocardial infarction, ischemic stroke, etc. Atherosclerotic plaque formation begins with the accumulation of foam cells and macrophages within the intima layer of the arterial wall. At the latter stage, the smooth muscle cells migrated from deeper artery wall layers, contributing to the fibrous cap formation and plaque stabilizing. A developed plaque gradually enters the lumen and narrows down the lumen to impede blood flow. We introduce a two-phase and macroscopic model to investigate the progression of plaque growth in its advanced stage and analyze the minimum gap (Lumen Clearance) within an atherosclerotic artery so that blood cells can pass through. Cardiac troponin, a high specificity and sensitivity biomarker, facilitates early detection of elevated myocardial infarction, Ischemic stroke, etc. risks. This study aims to establish a relationship between the troponin concentration in atherosclerotic arteries and their internal clearance, which could significantly improve our understanding of disease progression. Our observations show that the plaque undergoes rapid evolution in its initial stages, gradually slowing down over time to reach a steady state. At the same time, the lumen clearance exhibits an opposite behavior, decreasing slowly over time. Our study finds a positive correlation between plaque depth and troponin concentration in the blood and a negative relationship between troponin concentrations and lumen clearance in atherosclerotic arteries.
Many head pose estimation (HPE) methods promise the ability to create full-range datasets, theoretically allowing the estimation of the rotation and positioning of the head from various angles. However, these methods are only accurate within a range of head angles; exceeding this specific range led to significant inaccuracies. This is dominantly explained by unclear specificity of the coordinate systems and Euler Angles used in the foundational rotation matrix calculations. Here, we addressed these limitations by presenting (1) methods that accurately infer the correct coordinate system and Euler angles in the correct axis-sequence, (2) novel formulae for 2D geometric augmentations of the rotation matrices under the (SPECIFIC) coordinate system, (3) derivations for the correct drawing routines for rotation matrices and poses, and (4) mathematical experimentation and verification that allow proper pitch-yaw coverage for full-range head pose dataset generation. Performing our augmentation techniques to existing head pose estimation methods demonstrated a significant improvement to the model performance. Code will be released upon paper acceptance.
Neurodegeneration as measured through magnetic resonance imaging (MRI) is recognized as a potential biomarker for diagnosing Alzheimer's disease (AD), but is generally considered less specific than amyloid or tau based biomarkers. Due to a large amount of variability in brain anatomy between different individuals, we hypothesize that leveraging MRI time series can help improve specificity, by treating each patient as their own baseline. Here we turn to conditional variational autoencoders to generate individualized MRI predictions given the subject's age, disease status and one previous scan. Using serial imaging data from the Alzheimer's Disease Neuroimaging Initiative, we train a novel architecture to build a latent space distribution which can be sampled from to generate future predictions of changing anatomy. This enables us to extrapolate beyond the dataset and predict MRIs up to 10 years. We evaluated the model on a held-out set from ADNI and an independent dataset (from Open Access Series of Imaging Studies). By comparing to several alternatives, we show that our model produces more individualized images with higher resolution. Further, if an individual already has a follow-up MRI, we demonstrate a usage of our model to compute a likelihood ratio classifier for disease status. In practice, the model may be able to assist in early diagnosis of AD and provide a counterfactual baseline trajectory for treatment effect estimation. Furthermore, it generates a synthetic dataset that can potentially be used for downstream tasks such as anomaly detection and classification.
Calls to use open generative language models in academic research have highlighted the need for reproducibility and transparency in scientific research. However, the impact of generative AI extends well beyond academia, as corporations and public interest organizations have begun integrating these models into their data science pipelines. We expand this lens to include the impact of open models on organizations, focusing specifically on fact-checking organizations, which use AI to observe and analyze large volumes of circulating misinformation, yet must also ensure the reproducibility and impartiality of their work. We wanted to understand where fact-checking organizations use open models in their data science pipelines; what motivates their use of open models or proprietary models; and how their use of open or proprietary models can inform research on the societal impact of generative AI. To answer these questions, we conducted an interview study with N=24 professionals at 20 fact-checking organizations on six continents. Based on these interviews, we offer a five-component conceptual model of where fact-checking organizations employ generative AI to support or automate parts of their data science pipeline, including Data Ingestion, Data Analysis, Data Retrieval, Data Delivery, and Data Sharing. We then provide taxonomies of fact-checking organizations' motivations for using open models and the limitations that prevent them for further adopting open models, finding that they prefer open models for Organizational Autonomy, Data Privacy and Ownership, Application Specificity, and Capability Transparency. However, they nonetheless use proprietary models due to perceived advantages in Performance, Usability, and Safety, as well as Opportunity Costs related to participation in emerging generative AI ecosystems. Our work provides novel perspective on open models in data-driven organizations.
Sustainable aviation fuels have the potential for reducing emissions and environmental impact. To help identify viable sustainable aviation fuels and accelerate research, several machine learning models have been developed to predict relevant physiochemical properties. However, many of the models have limited applicability, leverage data from complex analytical techniques with confined spectral ranges, or use feature decomposition methods that have limited interpretability. Using liquid-phase Fourier Transform Infrared (FTIR) spectra, this study presents a structured method for creating accurate and interpretable property prediction models for neat molecules, aviation fuels, and blends. Liquid-phase FTIR spectra measurements can be collected quickly and consistently, offering high reliability, sensitivity, and component specificity using less than 2 mL of sample. The method first decomposes FTIR spectra into fundamental building blocks using Non-negative Matrix Factorization (NMF) to enable scientific analysis of FTIR spectra attributes and fuel properties. The NMF features are then used to create five ensemble models for predicting final boiling point, flash point, freezing point, density at 15C, and kinematic viscosity at -20C. All models were trained using experimental property data from neat molecules, aviation fuels, and blends. The models accurately predict properties while enabling interpretation of relationships between compositional elements of a fuel, such as functional groups or chemical classes, and its properties. To support sustainable aviation fuel research and development, the models and data are available on an interactive web tool.
The e-commerce platform has evolved rapidly due to its widespread popularity and convenience. Developing an e-commerce shopping assistant for customers is crucial to aiding them in quickly finding desired products and recommending precisely what they need. However, most previous shopping assistants face two main problems: (1) task-specificity, which necessitates the development of different models for various tasks, thereby increasing development costs and limiting effectiveness; and (2) poor generalization, where the trained model performs inadequately on up-to-date products. To resolve these issues, we employ Large Language Models (LLMs) to construct an omnipotent assistant, leveraging their adeptness at handling multiple tasks and their superior generalization capability. Nonetheless, LLMs lack inherent knowledge of e-commerce concepts. To address this, we create an instruction dataset comprising 65,000 samples and diverse tasks, termed as EshopInstruct. Through instruction tuning on our dataset, the assistant, named LLaSA, demonstrates the potential to function as an omnipotent assistant. Additionally, we propose various inference optimization strategies to enhance performance with limited inference resources. In the Amazon KDD Cup 2024 Challenge, our proposed method, LLaSA, achieved an overall ranking of 3rd place on ShopBench, including 57 tasks and approximately 20,000 questions, and we secured top-5 rankings in each track, especially in track4, where we achieved the best performance result among all student teams. Our extensive practices fully demonstrate that LLMs possess the great potential to be competent e-commerce shopping assistants.