Search

Multimodal/vision language models (VLMs) are increasingly being deployed in healthcare settings worldwide, necessitating robust benchmarks to ensure their safety, efficacy, and fairness. Multiple-choice question and answer (QA) datasets derived from national medical examinations have long served as valuable evaluation tools, but existing datasets are largely text-only and available in a limited subset of languages and countries. To address these challenges, we present WorldMedQA-V, an updated multilingual, multimodal benchmarking dataset designed to evaluate VLMs in healthcare. WorldMedQA-V includes 568 labeled multiple-choice QAs paired with 568 medical images from four countries (Brazil, Israel, Japan, and Spain), covering original languages and validated English translations by native clinicians, respectively. Baseline performance for common open- and closed-source models are provided in the local language and English translations, and with and without images provided to the model. The WorldMedQA-V benchmark aims to better match AI systems to the diverse healthcare environments in which they are deployed, fostering more equitable, effective, and representative applications., Comment: submitted for review, total of 14 pages

Non-Abelian gauge theories provide an accurate description of fundamental interactions, as both perturbation theory and quantum Monte Carlo computations in lattice gauge theory, it when applicable, show remarkable agreement with experimental data from particle colliders and cosmological observations. Complementing these computations, or combining them with quantum-inspired Hamiltonian lattice computations on quantum machines to improve continuum limit predictions with current quantum resources, is a formidable open challenge. Here, we propose a resource-efficient method to compute the running of the coupling in non-Abelian gauge theories beyond one spatial dimension. We first represent the Hamiltonian on periodic lattices in terms of loop variables and conjugate loop electric fields, exploiting the Gauss law to retain the gauge-independent ones. Then, we identify a local basis for small and large loops variationally to minimize the truncation error while computing the running of the coupling on small tori. Our method enables computations at arbitrary values of the bare coupling and lattice spacing with current quantum computers, simulators and tensor-network calculations, in regimes otherwise inaccessible., Comment: 12 pages + 17 pages appendix/references, 9 figures

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).

Medical decisions directly impact individuals' health and well-being. Extracting decision spans from clinical notes plays a crucial role in understanding medical decision-making processes. In this paper, we develop a new dataset called "MedDec", which contains clinical notes of eleven different phenotypes (diseases) annotated by ten types of medical decisions. We introduce the task of medical decision extraction, aiming to jointly extract and classify different types of medical decisions within clinical notes. We provide a comprehensive analysis of the dataset, develop a span detection model as a baseline for this task, evaluate recent span detection approaches, and employ a few metrics to measure the complexity of data samples. Our findings shed light on the complexities inherent in clinical decision extraction and enable future work in this area of research. The dataset and code are available through https://github.com/CLU-UML/MedDec., Comment: In Findings of the Association for Computational Linguistics ACL 2024

Algorithmic bias in healthcare mirrors existing data biases. However, the factors driving unfairness are not always known. Medical devices capture significant amounts of data but are prone to errors; for instance, pulse oximeters overestimate the arterial oxygen saturation of darker-skinned individuals, leading to worse outcomes. The impact of this bias in machine learning (ML) models remains unclear. This study addresses the technical challenges of quantifying the impact of medical device bias in downstream ML. Our experiments compare a "perfect world", without pulse oximetry bias, using SaO2 (blood-gas), to the "actual world", with biased measurements, using SpO2 (pulse oximetry). Under this counterfactual design, two models are trained with identical data, features, and settings, except for the method of measuring oxygen saturation: models using SaO2 are a "control" and models using SpO2 a "treatment". The blood-gas oximetry linked dataset was a suitable test-bed, containing 163,396 nearly-simultaneous SpO2 - SaO2 paired measurements, aligned with a wide array of clinical features and outcomes. We studied three classification tasks: in-hospital mortality, respiratory SOFA score in the next 24 hours, and SOFA score increase by two points. Models using SaO2 instead of SpO2 generally showed better performance. Patients with overestimation of O2 by pulse oximetry of > 3% had significant decreases in mortality prediction recall, from 0.63 to 0.59, P < 0.001. This mirrors clinical processes where biased pulse oximetry readings provide clinicians with false reassurance of patients' oxygen levels. A similar degradation happened in ML models, with pulse oximetry biases leading to more false negatives in predicting adverse outcomes., Comment: 10 pages; accepted at MICCAI's Third Workshop on Applications of Medical AI (2024)

Time-resolved three-dimensional flow MRI (4D flow MRI) provides a unique non-invasive solution to visualize and quantify hemodynamics in blood vessels such as the aortic arch. However, most current analysis methods for arterial 4D flow MRI use static artery walls because of the difficulty in obtaining a full cycle segmentation. To overcome this limitation, we propose a neural fields-based method that directly estimates continuous periodic wall deformations throughout the cardiac cycle. For a 3D + time imaging dataset, we optimize an implicit neural representation (INR) that represents a time-dependent velocity vector field (VVF). An ODE solver is used to integrate the VVF into a deformation vector field (DVF), that can deform images, segmentation masks, or meshes over time, thereby visualizing and quantifying local wall motion patterns. To properly reflect the periodic nature of 3D + time cardiovascular data, we impose periodicity in two ways. First, by periodically encoding the time input to the INR, and hence VVF. Second, by regularizing the DVF. We demonstrate the effectiveness of this approach on synthetic data with different periodic patterns, ECG-gated CT, and 4D flow MRI data. The obtained method could be used to improve 4D flow MRI analysis., Comment: 10 pages, 5 figures, STACOM 2024

The Cryogenic Underground Observatory for Rare Events (CUORE) is a detector array comprised by 988 5$\;$cm$\times$5$\;$cm$\times$5$\;$cm TeO$_2$ crystals held below 20 mK, primarily searching for neutrinoless double-beta decay in $^{130}$Te. Unprecedented in size amongst cryogenic calorimetric experiments, CUORE provides a promising setting for the study of exotic through-going particles. Using the first tonne-year of CUORE's exposure, we perform a search for hypothesized fractionally charged particles (FCPs), which are well-motivated by various Standard Model extensions and would have suppressed interactions with matter. No excess of FCP candidate tracks is observed over background, setting leading limits on the underground FCP flux with charges between $e/24-e/5$ at 90\% confidence level. Using the low background environment and segmented geometry of CUORE, we establish the sensitivity of tonne-scale sub-Kelvin detectors to diverse signatures of new physics., Comment: 7 pages, 5 figures

The deployment of large language models (LLMs) in healthcare has demonstrated substantial potential for enhancing clinical decision-making, administrative efficiency, and patient outcomes. However, the underrepresentation of diverse groups in the development and application of these models can perpetuate biases, leading to inequitable healthcare delivery. This paper presents a comprehensive scientometric analysis of LLM research for healthcare, including data from January 1, 2021, to July 1, 2024. By analyzing metadata from PubMed and Dimensions, including author affiliations, countries, and funding sources, we assess the diversity of contributors to LLM research. Our findings highlight significant gender and geographic disparities, with a predominance of male authors and contributions primarily from high-income countries (HICs). We introduce a novel journal diversity index based on Gini diversity to measure the inclusiveness of scientific publications. Our results underscore the necessity for greater representation in order to ensure the equitable application of LLMs in healthcare. We propose actionable strategies to enhance diversity and inclusivity in artificial intelligence research, with the ultimate goal of fostering a more inclusive and equitable future in healthcare innovation.

Medical knowledge is context-dependent and requires consistent reasoning across various natural language expressions of semantically equivalent phrases. This is particularly crucial for drug names, where patients often use brand names like Advil or Tylenol instead of their generic equivalents. To study this, we create a new robustness dataset, RABBITS, to evaluate performance differences on medical benchmarks after swapping brand and generic drug names using physician expert annotations. We assess both open-source and API-based LLMs on MedQA and MedMCQA, revealing a consistent performance drop ranging from 1-10\%. Furthermore, we identify a potential source of this fragility as the contamination of test data in widely used pre-training datasets. All code is accessible at https://github.com/BittermanLab/RABBITS, and a HuggingFace leaderboard is available at https://huggingface.co/spaces/AIM-Harvard/rabbits-leaderboard., Comment: submitted for review, total 15 pages

Large-scale multi-modal deep learning models have revolutionized domains such as healthcare, highlighting the importance of computational power. However, in resource-constrained regions like Low and Middle-Income Countries (LMICs), limited access to GPUs and data poses significant challenges, often leaving CPUs as the sole resource. To address this, we advocate for leveraging vector embeddings to enable flexible and efficient computational methodologies, democratizing multimodal deep learning across diverse contexts. Our paper investigates the efficiency and effectiveness of using vector embeddings from single-modal foundation models and multi-modal Vision-Language Models (VLMs) for multimodal deep learning in low-resource environments, particularly in healthcare. Additionally, we propose a simple yet effective inference-time method to enhance performance by aligning image-text embeddings. Comparing these approaches with traditional methods, we assess their impact on computational efficiency and model performance using metrics like accuracy, F1-score, inference time, training time, and memory usage across three medical modalities: BRSET (ophthalmology), HAM10000 (dermatology), and SatelliteBench (public health). Our findings show that embeddings reduce computational demands without compromising model performance. Furthermore, our alignment method improves performance in medical tasks. This research promotes sustainable AI practices by optimizing resources in constrained environments, highlighting the potential of embedding-based approaches for efficient multimodal learning. Vector embeddings democratize multimodal deep learning in LMICs, particularly in healthcare, enhancing AI adaptability in varied use cases.

We present the model we developed to reconstruct the CUORE radioactive background based on the analysis of an experimental exposure of 1038.4 kg yr. The data reconstruction relies on a simultaneous Bayesian fit applied to energy spectra over a broad energy range. The high granularity of the CUORE detector, together with the large exposure and extended stable operations, allow for an in-depth exploration of both spatial and time dependence of backgrounds. We achieve high sensitivity to both bulk and surface activities of the materials of the setup, detecting levels as low as 10 nBq kg$^{-1}$ and 0.1 nBq cm$^{-2}$, respectively. We compare the contamination levels we extract from the background model with prior radio-assay data, which informs future background risk mitigation strategies. The results of this background model play a crucial role in constructing the background budget for the CUPID experiment as it will exploit the same CUORE infrastructure.

The ethical integration of Artificial Intelligence (AI) in healthcare necessitates addressing fairness-a concept that is highly context-specific across medical fields. Extensive studies have been conducted to expand the technical components of AI fairness, while tremendous calls for AI fairness have been raised from healthcare. Despite this, a significant disconnect persists between technical advancements and their practical clinical applications, resulting in a lack of contextualized discussion of AI fairness in clinical settings. Through a detailed evidence gap analysis, our review systematically pinpoints several deficiencies concerning both healthcare data and the provided AI fairness solutions. We highlight the scarcity of research on AI fairness in many medical domains where AI technology is increasingly utilized. Additionally, our analysis highlights a substantial reliance on group fairness, aiming to ensure equality among demographic groups from a macro healthcare system perspective; in contrast, individual fairness, focusing on equity at a more granular level, is frequently overlooked. To bridge these gaps, our review advances actionable strategies for both the healthcare and AI research communities. Beyond applying existing AI fairness methods in healthcare, we further emphasize the importance of involving healthcare professionals to refine AI fairness concepts and methods to ensure contextually relevant and ethically sound AI applications in healthcare.

The current experiments searching for neutrinoless double-$\beta$ ($0\nu\beta\beta$) decay also collect large statistics of Standard Model allowed two-neutrino double-$\beta$ ($2\nu\beta\beta$) decay events. These can be used to search for Beyond Standard Model (BSM) physics via $2\nu\beta\beta$ decay spectral distortions. $^{100}$Mo has a natural advantage due to its relatively short half-life, allowing higher $2\nu\beta\beta$ decay statistics at equal exposures compared to the other isotopes. We demonstrate the potential of the dual read-out bolometric technique exploiting a $^{100}$Mo exposure of 1.47 kg $\times$ y, acquired in the CUPID-Mo experiment at the Modane underground laboratory (France). We set limits on $0\nu\beta\beta$ decays with the emission of one or more Majorons, on $2\nu\beta\beta$ decay with Lorentz violation, and $2\nu\beta\beta$ decay with a sterile neutrino emission. In this analysis, we investigate the systematic uncertainty induced by modeling the $2\nu\beta\beta$ decay spectral shape parameterized through an improved model, an effect never considered before. This work motivates searches for BSM processes in the upcoming CUPID experiment, which will collect the largest amount of $2\nu\beta\beta$ decay events among the next-generation experiments.

Large language models (LLMs) are increasingly essential in processing natural languages, yet their application is frequently compromised by biases and inaccuracies originating in their training data. In this study, we introduce Cross-Care, the first benchmark framework dedicated to assessing biases and real world knowledge in LLMs, specifically focusing on the representation of disease prevalence across diverse demographic groups. We systematically evaluate how demographic biases embedded in pre-training corpora like $ThePile$ influence the outputs of LLMs. We expose and quantify discrepancies by juxtaposing these biases against actual disease prevalences in various U.S. demographic groups. Our results highlight substantial misalignment between LLM representation of disease prevalence and real disease prevalence rates across demographic subgroups, indicating a pronounced risk of bias propagation and a lack of real-world grounding for medical applications of LLMs. Furthermore, we observe that various alignment methods minimally resolve inconsistencies in the models' representation of disease prevalence across different languages. For further exploration and analysis, we make all data and a data visualization tool available at: www.crosscare.net., Comment: Submitted for review, data visualization tool available at: www.crosscare.net

Background Advancements in Large Language Models (LLMs) hold transformative potential in healthcare, however, recent work has raised concern about the tendency of these models to produce outputs that display racial or gender biases. Although training data is a likely source of such biases, exploration of disease and demographic associations in text data at scale has been limited. Methods We conducted a large-scale textual analysis using a dataset comprising diverse web sources, including Arxiv, Wikipedia, and Common Crawl. The study analyzed the context in which various diseases are discussed alongside markers of race and gender. Given that LLMs are pre-trained on similar datasets, this approach allowed us to examine the potential biases that LLMs may learn and internalize. We compared these findings with actual demographic disease prevalence as well as GPT-4 outputs in order to evaluate the extent of bias representation. Results Our findings indicate that demographic terms are disproportionately associated with specific disease concepts in online texts. gender terms are prominently associated with disease concepts, while racial terms are much less frequently associated. We find widespread disparities in the associations of specific racial and gender terms with the 18 diseases analyzed. Most prominently, we see an overall significant overrepresentation of Black race mentions in comparison to population proportions. Conclusions Our results highlight the need for critical examination and transparent reporting of biases in LLM pretraining datasets. Our study suggests the need to develop mitigation strategies to counteract the influence of biased training data in LLMs, particularly in sensitive domains such as healthcare.

In the big data era, integrating diverse data modalities poses significant challenges, particularly in complex fields like healthcare. This paper introduces a new process model for multimodal Data Fusion for Data Mining, integrating embeddings and the Cross-Industry Standard Process for Data Mining with the existing Data Fusion Information Group model. Our model aims to decrease computational costs, complexity, and bias while improving efficiency and reliability. We also propose "disentangled dense fusion", a novel embedding fusion method designed to optimize mutual information and facilitate dense inter-modality feature interaction, thereby minimizing redundant information. We demonstrate the model's efficacy through three use cases: predicting diabetic retinopathy using retinal images and patient metadata, domestic violence prediction employing satellite imagery, internet, and census data, and identifying clinical and demographic features from radiography images and clinical notes. The model achieved a Macro F1 score of 0.92 in diabetic retinopathy prediction, an R-squared of 0.854 and sMAPE of 24.868 in domestic violence prediction, and a macro AUC of 0.92 and 0.99 for disease prediction and sex classification, respectively, in radiological analysis. These results underscore the Data Fusion for Data Mining model's potential to significantly impact multimodal data processing, promoting its adoption in diverse, resource-constrained settings., Comment: 6 figures, 5 tables

The matter-antimatter asymmetry underlines the incompleteness of the current understanding of particle physics. Neutrinoless double-beta ($0\nu \beta\beta$) decay may help explain this asymmetry, while unveiling the Majorana nature of the neutrino. The CUORE experiment searches for $0\nu \beta\beta$ decay of $^{130}$Te using a tonne-scale cryogenic calorimeter operated at milli-kelvin temperatures. We report no evidence for $0\nu \beta\beta$ decay and place a lower limit on the half-life of T$_{1/2}$ $>$ 3.8 $\times$ 10$^{25}$ years (90% C.I.) with over 2 tonne$\cdot$year TeO$_2$ exposure. The tools and techniques developed for this result and the 5 year stable operation of nearly 1000 detectors demonstrate the infrastructure for a next-generation experiment capable of searching for $0\nu \beta\beta$ decay across multiple isotopes.

Objective: Randomized controlled trial (RCT) results often inform clinical decision-making, but the highly curated populations of trials and the care provided during the trial are often not reflective of real-world practice. The objective of this scoping review is to identify the ability of methods to transport findings from RCTs to target populations. Study design: A scoping review was conducted on the literature focusing on the transportability of the results from RCTs to observational cohorts. Each study was assessed based on the methodology used for transportability and the extent to which the treatment effect from the RCT was estimated in the target population in observational data. Results: A total of 15 published papers were included. The research topics include cardiovascular diseases, infectious diseases, psychiatry, oncology, orthopedics, anesthesiology, and hematology. These studies show that the findings from RCTs could be translated to real-world settings, with varying degrees of effect size and precision. In some cases, the estimated treatment effect for the target population were statistically significantly different from those in RCTs. Conclusion: Despite variations in the magnitude of effects between RCTs and real-world studies, transportability methods play an important role in effectively bridging the RCTs and real-world care delivery, offering valuable insights for evidence-based medicine.

Current bounds on neutrino Majorana mass are affected by significant uncertainties in the nuclear calculations for neutrinoless double-beta decay. A key issue for a data-driven improvement of the nuclear theory is the actual value of the axial coupling constant $g_A$, which can be investigated through forbidden $\beta$-decays. We present the first measurement of 4$^{th}$-forbidden $\beta$-decay of $^{115}$In with a cryogenic calorimeter based on Indium Iodide. Exploiting the enhanced spectral shape method for the first time to this isotope, our study accurately determines simultaneously spectral shape, $g_A$, and half-life. The Interacting Shell Model, which best fits our data, indicates a half-life for this decay at $T_{1/2}=(5.26\pm0.06) \times 10^{14}$,yr.

Dengue fever presents a substantial challenge in developing countries where sanitation infrastructure is inadequate. The absence of comprehensive healthcare systems exacerbates the severity of dengue infections, potentially leading to life-threatening circumstances. Rapid response to dengue outbreaks is also challenging due to limited information exchange and integration. While timely dengue outbreak forecasts have the potential to prevent such outbreaks, the majority of dengue prediction studies have predominantly relied on data that impose significant burdens on individual countries for collection. In this study, our aim is to improve health equity in resource-constrained countries by exploring the effectiveness of high-resolution satellite imagery as a nontraditional and readily accessible data source. By leveraging the wealth of publicly available and easily obtainable satellite imagery, we present a scalable satellite extraction framework based on Sentinel Hub, a cloud-based computing platform. Furthermore, we introduce DengueNet, an innovative architecture that combines Vision Transformer, Radiomics, and Long Short-term Memory to extract and integrate spatiotemporal features from satellite images. This enables dengue predictions on an epi-week basis. To evaluate the effectiveness of our proposed method, we conducted experiments on five municipalities in Colombia. We utilized a dataset comprising 780 high-resolution Sentinel-2 satellite images for training and evaluation. The performance of DengueNet was assessed using the mean absolute error (MAE) metric. Across the five municipalities, DengueNet achieved an average MAE of 43.92. Our findings strongly support the efficacy of satellite imagery as a valuable resource for dengue prediction, particularly in informing public health policies within countries where manually collected data is scarce and dengue virus prevalence is severe., Comment: Published at the IJCAI 2023 Workshop on Bridge-AI: from Climate Change to Health Equity (BridgeAICCHE)., Macao, S.A.R

Diabetic retinopathy (DR) is a prevalent complication of diabetes associated with a significant risk of vision loss. Timely identification is critical to curb vision impairment. Algorithms for DR staging from digital fundus images (DFIs) have been recently proposed. However, models often fail to generalize due to distribution shifts between the source domain on which the model was trained and the target domain where it is deployed. A common and particularly challenging shift is often encountered when the source- and target-domain supports do not fully overlap. In this research, we introduce DRStageNet, a deep learning model designed to mitigate this challenge. We used seven publicly available datasets, comprising a total of 93,534 DFIs that cover a variety of patient demographics, ethnicities, geographic origins and comorbidities. We fine-tune DINOv2, a pretrained model of self-supervised vision transformer, and implement a multi-source domain fine-tuning strategy to enhance generalization performance. We benchmark and demonstrate the superiority of our method to two state-of-the-art benchmarks, including a recently published foundation model. We adapted the grad-rollout method to our regression task in order to provide high-resolution explainability heatmaps. The error analysis showed that 59\% of the main errors had incorrect reference labels. DRStageNet is accessible at URL [upon acceptance of the manuscript].

In the era of data-driven economies, incentive systems and loyalty programs, have become ubiquitous in various sectors, including advertising, retail, travel, and financial services. While these systems offer advantages for both users and companies, they necessitate the transfer and analysis of substantial amounts of sensitive data. Privacy concerns have become increasingly pertinent, necessitating the development of privacy-preserving incentive protocols. Despite the rising demand for secure and decentralised systems, the existing landscape lacks a comprehensive solution. In this work, we propose the BOOMERANG protocol, a novel decentralised privacy-preserving incentive protocol that leverages cryptographic black box accumulators to securely and privately store user interactions within the incentive system. Moreover, the protocol employs zero-knowledge proofs to transparently compute rewards for users, ensuring verifiability while preserving their privacy. To further enhance public verifiability and transparency, we utilise a smart contract on a Layer 1 blockchain to verify these zero-knowledge proofs. The careful combination of black box accumulators and zero-knowledge proofs makes the BOOMERANG protocol highly efficient., Comment: updated to latest version

We study the effects of the first nontrivial hexaquark, $d^*$(2380), on the equation of state of dense neutron star matter and investigate the consequences of its existence for neutron stars. The matter in the core regions of neutron stars is described using density-dependent relativistic mean-field theory. Our results show that within the parameter spaces examined in our paper, (i) the critical density at which the $d^*$ condensate emerges lies between 4 and 5 times the nuclear saturation density, (ii) $d^*$ hexaquarks are found to exist only in rather massive neutron stars, (iii) only relatively small fractions of the matter in the core of a massive neutron star may contain hexaquarks., Comment: 15 pages, 7 figures. Accepted for publication in Physical Review D

Clinical AI model reporting cards should be expanded to incorporate a broad bias reporting of both social and non-social factors. Non-social factors consider the role of other factors, such as disease dependent, anatomic, or instrument factors on AI model bias, which are essential to ensure safe deployment.

Over the past few years, research has witnessed the advancement of deep learning models trained on large datasets, some even encompassing millions of examples. While these impressive performance on their hidden test sets, they often underperform when assessed on external datasets. Recognizing the critical role of generalization in medical AI development, many prestigious journals now require reporting results both on the local hidden test set as well as on external datasets before considering a study for publication. Effectively, the field of medical AI has transitioned from the traditional usage of a single dataset that is split into train and test to a more comprehensive framework using multiple datasets, some of which are used for model development (source domain) and others for testing (target domains). However, this new experimental setting does not necessarily resolve the challenge of generalization. This is because of the variability encountered in intended use and specificities across hospital cultures making the idea of universally generalizable systems a myth. On the other hand, the systematic, and a fortiori recurrent re-calibration, of models at the individual hospital level, although ideal, may be overoptimistic given the legal, regulatory and technical challenges that are involved. Re-calibration using transfer learning may not even be possible in some instances where reference labels of target domains are not available. In this perspective we establish a hierarchical three-level scale system reflecting the generalization level of a medical AI algorithm. This scale better reflects the diversity of real-world medical scenarios per which target domain data for re-calibration of models may or not be available and if it is, may or not have reference labels systematically available.

In this Perspective article we report on recent progress on studies of synthetic dimensions, mostly, but not only, based on the research realized around the Barcelona groups (ICFO, UAB), Donostia (DIPC), Pozna\'n (UAM), Krak\'ow (UJ), and Allahabad (HRI). The concept of synthetic dimensions works particularly well in atomic physics, quantum optics, and photonics, where the internal degrees of freedom (Zeeman sublevels of the ground state, metastable excited states, or motional states for atoms, and angular momentum states or transverse modes for photons) provide the synthetic space. We describe our attempts to design quantum simulators with synthetic dimensions, to mimic curved spaces, artificial gauge fields, lattice gauge theories, twistronics, quantum random walks, and more.

Many real-world image recognition problems, such as diagnostic medical imaging exams, are "long-tailed" $\unicode{x2013}$ there are a few common findings followed by many more relatively rare conditions. In chest radiography, diagnosis is both a long-tailed and multi-label problem, as patients often present with multiple findings simultaneously. While researchers have begun to study the problem of long-tailed learning in medical image recognition, few have studied the interaction of label imbalance and label co-occurrence posed by long-tailed, multi-label disease classification. To engage with the research community on this emerging topic, we conducted an open challenge, CXR-LT, on long-tailed, multi-label thorax disease classification from chest X-rays (CXRs). We publicly release a large-scale benchmark dataset of over 350,000 CXRs, each labeled with at least one of 26 clinical findings following a long-tailed distribution. We synthesize common themes of top-performing solutions, providing practical recommendations for long-tailed, multi-label medical image classification. Finally, we use these insights to propose a path forward involving vision-language foundation models for few- and zero-shot disease classification., Comment: Update after major revision

Over the past two decades, exponential growth in data availability, computational power, and newly available modeling techniques has led to an expansion in interest, investment, and research in Artificial Intelligence (AI) applications. Ophthalmology is one of many fields that seek to benefit from AI given the advent of telemedicine screening programs and the use of ancillary imaging. However, before AI can be widely deployed, further work must be done to avoid the pitfalls within the AI lifecycle. This review article breaks down the AI lifecycle into seven steps: data collection; defining the model task; data pre-processing and labeling; model development; model evaluation and validation; deployment; and finally, post-deployment evaluation, monitoring, and system recalibration and delves into the risks for harm at each step and strategies for mitigating them.