Using generative AI to investigate medical imagery models and datasets.

BACKGROUND: AI models have shown promise in performing many medical imaging tasks. However, our ability to explain what signals these models have learned is severely lacking. Explanations are needed in order to increase the trust of doctors in AI-based models, especially in domains where AI prediction capabilities surpass those of humans. Moreover, such explanations could enable novel scientific discovery by uncovering signals in the data that aren't yet known to experts. METHODS: In this paper, we present a workflow for generating hypotheses to understand which visual signals in images are correlated with a classification model's predictions for a given task. This approach leverages an automatic visual explanation algorithm followed by interdisciplinary expert review. We propose the following 4 steps: (i) Train a classifier to perform a given task to assess whether the imagery indeed contains signals relevant to the task; (ii) Train a StyleGAN-based image generator with an architecture that enables guidance by the classifier ("StylEx"); (iii) Automatically detect, extract, and visualize the top visual attributes that the classifier is sensitive towards. For visualization, we independently modify each of these attributes to generate counterfactual visualizations for a set of images (i.e., what the image would look like with the attribute increased or decreased); (iv) Formulate hypotheses for the underlying mechanisms, to stimulate future research. Specifically, present the discovered attributes and corresponding counterfactual visualizations to an interdisciplinary panel of experts so that hypotheses can account for social and structural determinants of health (e.g., whether the attributes correspond to known patho-physiological or socio-cultural phenomena, or could be novel discoveries). FINDINGS: To demonstrate the broad applicability of our approach, we present results on eight prediction tasks across three medical imaging modalities-retinal fundus photographs, external eye photographs, and chest radiographs. We showcase examples where many of the automatically-learned attributes clearly capture clinically known features (e.g., types of cataract, enlarged heart), and demonstrate automatically-learned confounders that arise from factors beyond physiological mechanisms (e.g., chest X-ray underexposure is correlated with the classifier predicting abnormality, and eye makeup is correlated with the classifier predicting low hemoglobin levels). We further show that our method reveals a number of physiologically plausible, previously-unknown attributes based on the literature (e.g., differences in the fundus associated with self-reported sex, which were previously unknown). INTERPRETATION: Our approach enables hypotheses generation via attribute visualizations and has the potential to enable researchers to better understand, improve their assessment, and extract new knowledge from AI-based models, as well as debug and design better datasets. Though not designed to infer causality, importantly, we highlight that attributes generated by our framework can capture phenomena beyond physiology or pathophysiology, reflecting the real world nature of healthcare delivery and socio-cultural factors, and hence interdisciplinary perspectives are critical in these investigations. Finally, we will release code to help researchers train their own StylEx models and analyze their predictive tasks of interest, and use the methodology presented in this paper for responsible interpretation of the revealed attributes. FUNDING: Google.

Alignment of brain embeddings and artificial contextual embeddings in natural language points to common geometric patterns.

Contextual embeddings, derived from deep language models (DLMs), provide a continuous vectorial representation of language. This embedding space differs fundamentally from the symbolic representations posited by traditional psycholinguistics. We hypothesize that language areas in the human brain, similar to DLMs, rely on a continuous embedding space to represent language. To test this hypothesis, we densely record the neural activity patterns in the inferior frontal gyrus (IFG) of three participants using dense intracranial arrays while they listened to a 30-minute podcast. From these fine-grained spatiotemporal neural recordings, we derive a continuous vectorial representation for each word (i.e., a brain embedding) in each patient. Using stringent zero-shot mapping we demonstrate that brain embeddings in the IFG and the DLM contextual embedding space have common geometric patterns. The common geometric patterns allow us to predict the brain embedding in IFG of a given left-out word based solely on its geometrical relationship to other non-overlapping words in the podcast. Furthermore, we show that contextual embeddings capture the geometry of IFG embeddings better than static word embeddings. The continuous brain embedding space exposes a vector-based neural code for natural language processing in the human brain.

Global prediction of extreme floods in ungauged watersheds.

Floods are one of the most common natural disasters, with a disproportionate impact in developing countries that often lack dense streamflow gauge networks1. Accurate and timely warnings are critical for mitigating flood risks2, but hydrological simulation models typically must be calibrated to long data records in each watershed. Here we show that artificial intelligence-based forecasting achieves reliability in predicting extreme riverine events in ungauged watersheds at up to a five-day lead time that is similar to or better than the reliability of nowcasts (zero-day lead time) from a current state-of-the-art global modelling system (the Copernicus Emergency Management Service Global Flood Awareness System). In addition, we achieve accuracies over five-year return period events that are similar to or better than current accuracies over one-year return period events. This means that artificial intelligence can provide flood warnings earlier and over larger and more impactful events in ungauged basins. The model developed here was incorporated into an operational early warning system that produces publicly available (free and open) forecasts in real time in over 80 countries. This work highlights a need for increasing the availability of hydrological data to continue to improve global access to reliable flood warnings.

A neural encoder for earthquake rate forecasting.

Forecasting the timing of earthquakes is a long-standing challenge. Moreover, it is still debated how to formulate this problem in a useful manner, or to compare the predictive power of different models. Here, we develop a versatile neural encoder of earthquake catalogs, and apply it to the fundamental problem of earthquake rate prediction, in the spatio-temporal point process framework. The epidemic type aftershock sequence model (ETAS) effectively learns a small number of parameters to constrain the assumed functional forms for the space and time correlations of earthquake sequences (e.g., Omori-Utsu law). Here we introduce learned spatial and temporal embeddings for point process earthquake forecasting models that capture complex correlation structures. We demonstrate the generality of this neural representation as compared with ETAS model using train-test data splits and how it enables the incorporation additional geophysical information. In rate prediction tasks, the generalized model shows [Formula: see text] improvement in information gain per earthquake and the simultaneous learning of anisotropic spatial structures analogous to fault traces. The trained network can be also used to perform short-term prediction tasks, showing similar improvement while providing a 1000-fold reduction in run-time.

Caravan - A global community dataset for large-sample hydrology.

High-quality datasets are essential to support hydrological science and modeling. Several CAMELS (Catchment Attributes and Meteorology for Large-sample Studies) datasets exist for specific countries or regions, however these datasets lack standardization, which makes global studies difficult. This paper introduces a dataset called Caravan (a series of CAMELS) that standardizes and aggregates seven existing large-sample hydrology datasets. Caravan includes meteorological forcing data, streamflow data, and static catchment attributes (e.g., geophysical, sociological, climatological) for 6830 catchments. Most importantly, Caravan is both a dataset and open-source software that allows members of the hydrology community to extend the dataset to new locations by extracting forcing data and catchment attributes in the cloud. Our vision is for Caravan to democratize the creation and use of globally-standardized large-sample hydrology datasets. Caravan is a truly global open-source community resource.

Shared computational principles for language processing in humans and deep language models.

Departing from traditional linguistic models, advances in deep learning have resulted in a new type of predictive (autoregressive) deep language models (DLMs). Using a self-supervised next-word prediction task, these models generate appropriate linguistic responses in a given context. In the current study, nine participants listened to a 30-min podcast while their brain responses were recorded using electrocorticography (ECoG). We provide empirical evidence that the human brain and autoregressive DLMs share three fundamental computational principles as they process the same natural narrative: (1) both are engaged in continuous next-word prediction before word onset; (2) both match their pre-onset predictions to the incoming word to calculate post-onset surprise; (3) both rely on contextual embeddings to represent words in natural contexts. Together, our findings suggest that autoregressive DLMs provide a new and biologically feasible computational framework for studying the neural basis of language.

Active deep learning to detect demographic traits in free-form clinical notes.

The free-form portions of clinical notes are a significant source of information for research, but before they can be used, they must be de-identified to protect patients' privacy. De-identification efforts have focused on known identifier types (names, ages, dates, addresses, ID's, etc.). However, a note can contain residual "Demographic Traits" (DTs), unique enough to re-identify the patient when combined with other such facts. Here we examine whether any residual risks remain after removing these identifiers. After manually annotating over 140,000 words worth of medical notes, we found no remaining directly identifying information, and a low prevalence of demographic traits, such as marital status or housing type. We developed an annotation guide to the discovered Demographic Traits (DTs) and used it to label MIMIC-III and i2b2-2006 clinical notes as test sets. We then designed a "bootstrapped" active learning iterative process for identifying DTs: we tentatively labeled as positive all sentences in the DT-rich note sections, used these to train a binary classifier, manually corrected acute errors, and retrained the classifier. This train-and-correct process may be iterated. Our active learning process significantly improved the classifier's accuracy. Moreover, our BERT-based model outperformed non-neural models when trained on both tentatively labeled data and manually relabeled examples. To facilitate future research and benchmarking, we also produced and made publicly available our human annotated DT-tagged datasets. We conclude that directly identifying information is virtually non-existent in the multiple medical note types we investigated. Demographic traits are present in medical notes, but can be detected with high accuracy using a cost-effective human-in-the-loop active learning process, and redacted if desired.2.

Customization scenarios for de-identification of clinical notes.

BACKGROUND: Automated machine-learning systems are able to de-identify electronic medical records, including free-text clinical notes. Use of such systems would greatly boost the amount of data available to researchers, yet their deployment has been limited due to uncertainty about their performance when applied to new datasets. OBJECTIVE: We present practical options for clinical note de-identification, assessing performance of machine learning systems ranging from off-the-shelf to fully customized. METHODS: We implement a state-of-the-art machine learning de-identification system, training and testing on pairs of datasets that match the deployment scenarios. We use clinical notes from two i2b2 competition corpora, the Physionet Gold Standard corpus, and parts of the MIMIC-III dataset. RESULTS: Fully customized systems remove 97-99% of personally identifying information. Performance of off-the-shelf systems varies by dataset, with performance mostly above 90%. Providing a small labeled dataset or large unlabeled dataset allows for fine-tuning that improves performance over off-the-shelf systems. CONCLUSION: Health organizations should be aware of the levels of customization available when selecting a de-identification deployment solution, in order to choose the one that best matches their resources and target performance level.

Assigning Israeli medical graduates to internships.

BACKGROUND: Physicians in Israel are required to do an internship in an accredited hospital upon completion of the medical studies, and prior to receiving the medical license. For most students, the assignment is determined by a lottery, which takes into consideration the preferences of these students. OBJECTIVES: We propose a novel way to perform this lottery, in which (on average) a larger number of students gets one of their top choices. We report about implementing this method in the 2014 Internship Lottery in Israel. METHODS: The new method is based on calculating a tentative lottery, in which each student has some probability of getting to each hospital. Then a computer program "trades" between the students, where trade is performed only if it is beneficial to both sides. This trade creates surplus, which translates to more students getting one of their top choices. RESULTS: The average student improved his place by 0.91 seats. CONCLUSIONS: The new method can improve the welfare of medical graduates, by giving them more probability to get to one of their top choices. It can be applied in internship markets in other countries as well.

Quantum state restoration and single-copy tomography for ground states of Hamiltonians.

Given a single copy of an unknown quantum state, the no-cloning theorem limits the amount of information that can be extracted from it. Given a gapped Hamiltonian, in most situations it is impractical to compute properties of its ground state, even though in principle all the information about the ground state is encoded in the Hamiltonian. We show in this Letter that if you know the Hamiltonian of a system and have a single copy of its ground state, you can use a quantum computer to efficiently compute its local properties. Specifically, in this scenario, we give efficient algorithms that copy small subsystems of the state and estimate the full statistics of any local measurement.

Quantum algorithm for linear systems of equations.

Solving linear systems of equations is a common problem that arises both on its own and as a subroutine in more complex problems: given a matrix A and a vector b(-->), find a vector x(-->) such that Ax(-->) = b(-->). We consider the case where one does not need to know the solution x(-->) itself, but rather an approximation of the expectation value of some operator associated with x(-->), e.g., x(-->)(dagger) Mx(-->) for some matrix M. In this case, when A is sparse, N x N and has condition number kappa, the fastest known classical algorithms can find x(-->) and estimate x(-->)(dagger) Mx(-->) in time scaling roughly as N square root(kappa). Here, we exhibit a quantum algorithm for estimating x(-->)(dagger) Mx(-->) whose runtime is a polynomial of log(N) and kappa. Indeed, for small values of kappa [i.e., poly log(N)], we prove (using some common complexity-theoretic assumptions) that any classical algorithm for this problem generically requires exponentially more time than our quantum algorithm.