Scientific causal inference is frequently data limited: a target study may contain only tens of randomized or quasi-experimental observations, while related domains contain much larger observational and interventional datasets. We investigate whether intervention-rich cross-task pretraining can provide a reusable causal prior that improves treatment-effect estimation in a new small-data domain. We introduce SciCFM, an episodic causal pretraining framework that encodes support sets from heterogeneous structural causal models and adapts the learned mechanism representation to a target dose-response task. Three simulation studies were conducted. First, under changing hidden confounding, outcome-related selection, overlap restriction, measurement distortion, and nonlinear response mechanisms, causal pretraining reduced mean precision in estimation of heterogeneous effect (PEHE) by 42.8–53.8% relative to observational pretraining. Second, in a low-dimensional target regime, a directly fitted ridge learner outperformed SciCFM, demonstrating that pretraining is not universally advantageous. Third, in a harder regime with 30 covariates, six sparse active variables, latent response subgroups, and nonlinear threshold, saturation, interaction, and piecewise effects, SciCFM reduced PEHE relative to the strongest direct learner by 29.4% at n=16, 14.6% at n=32, and 3.2% at n=64. These pilot results support a regime-dependent conclusion: causal pretraining is most useful when target samples are extremely small and causal mechanisms are complex but recur across tasks; simple correctly regularized estimators remain preferable when the target mechanism is low dimensional.

Machine learning models play a vital role in making predictions and deriving insights from data and are being increasingly used for causal inference. To preserve user privacy, it is important to enable the model to forget some of its learning/captured information about a given user (machine unlearning). This paper introduces the concept of machine unlearning for causal inference, particularly propensity score matching and treatment effect estimation, which aims to refine and improve the performance of machine learning models for causal analysis given the above unlearning requirements. The paper presents a methodology for machine unlearning using a neural network-based propensity score model. The dataset used in the study is the Lalonde dataset, a widely used dataset for evaluating the effectiveness i.e. the treatment effect of job training programs. The methodology involves training an initial propensity score model on the original dataset and then creating forget sets by selectively removing instances, as well as matched instance pairs. based on propensity score matching. These forget sets are used to evaluate the retrained model, allowing for the elimination of unwanted associations. The actual retraining of the model is performed using the retain set. The experimental results demonstrate the effectiveness of the machine unlearning approach. The distribution and histogram analysis of propensity scores before and after unlearning provide insights into the impact of the unlearning process on the data. This study represents the first attempt to apply machine unlearning techniques to causal inference.

This paper explores the use of machine learning (ML) for predictive modeling on clinical data from the EMory BrEast imaging Dataset (EMBED) [1] with a focus on ML model fairness analysis. The aim of this study is to develop and evaluate fair machine learning models that can accurately predict breast cancer risk. We trained and tested various machine-learning models. Our findings show that machine learning can be effective for predicting breast cancer risk or diagnosing breast cancer, and that fairness considerations are crucial in the development of such models. Overall, our study highlights the potential of machine learning for clinical applications while emphasizing the need for ethical and fair practices in this field.

The paper presents a time series analysis of the air, water and soil data from the East Palestine, Ohio, train derailment site. The goal of the analysis is to investigate the temporal pattern of chemical pollutant levels in air, water and soil and to build machine learning and statistical models for forecasting various chemical concentrations over time.

In this paper, we use deep learning techniques to segment different regions from breast cancer histopathology images, such as tumor nucleus, epithelium and stromal areas. Then, in the second stage, the deep segmentation features learned by the neural network are used to predict individual patient survival, using random forest based classification. We show that the deep segmentation network features can predict survival very well, and outperform classical computer vision based shape, texture and other feature descriptors used in earlier research for the same survival prediction task.

The length of hospital stay (LOS) and the type of discharge are important indicators of how well care is provided at a hospital. The purpose of this study is to leverage in-patient data collected at the hospital to help determine the factors that influence the length of hospital stay and type of discharge. Our research focuses on estimating if the person survived or not after they were admitted to the hospital, as well as the type of discharge. The study uses a retrospective design and examines information from hospital discharged patients’ medical records. Demographic information, diagnosis, treatment, and discharge status were included in the data. We have used the PEDALFAST dataset which stands for PEDiatric Validation of Variables in Trauma. A survey of patients to find out how they feel about the quality of care they received while they were in the hospital was also a part of the study dataset. The findings of this study will shed light on the ways in which various factors influence the LOS in the hospital and the type of discharge, assisting in the formulation of strategies to enhance the quality and effectiveness of health care delivery.

CONTEXT.—:Breast carcinoma grade, as determined by the Nottingham Grading System (NGS), is an important criterion for determining prognosis. The NGS is based on 3 parameters: tubule formation (TF), nuclear pleomorphism (NP), and mitotic count (MC). The advent of digital pathology and artificial intelligence (AI) have increased interest in virtual microscopy using digital whole slide imaging (WSI) more broadly.OBJECTIVE.—:To compare concordance in breast carcinoma grading between AI and a multi-institutional group of breast pathologists using digital WSI.DESIGN.—:We have developed an automated NGS framework using deep learning. Six pathologists and AI independently reviewed a digitally scanned slide from 137 invasive carcinomas and assigned a grade based on scoring of the TF, NP, and MC.RESULTS.—:Interobserver agreement for the pathologists and AI for overall grade was moderate (κ = 0.471). Agreement was good (κ = 0.681), moderate (κ = 0.442), and fair (κ = 0.368) for grades 1, 3, and 2, respectively. Observer pair concordance for AI and individual pathologists ranged from fair to good (κ = 0.313-0.606). Perfect agreement was observed in 25 cases (27.4%). Interobserver agreement for the individual components was best for TF (κ = 0.471 each) followed by NP (κ = 0.342) and was worst for MC (κ = 0.233). There were no observed differences in concordance amongst pathologists alone versus pathologists + AI.CONCLUSIONS.—:Ours is the first study comparing concordance in breast carcinoma grading between a multi-institutional group of pathologists using virtual microscopy to a newly developed WSI AI methodology. Using explainable methods, AI demonstrated similar concordance to pathologists alone.

Detecting various types of cells in and around the tumor matrix holds a special significance in characterizing the tumor micro-environment for cancer prognostication and research. Automating the tasks of detecting, segmenting, and classifying nuclei can free up the pathologists' time for higher value tasks and reduce errors due to fatigue and subjectivity. To encourage the computer vision research community to develop and test algorithms for these tasks, we prepared a large and diverse dataset of nucleus boundary annotations and class labels. The dataset has over 46,000 nuclei from 37 hospitals, 71 patients, four organs, and four nucleus types. We also organized a challenge around this dataset as a satellite event at the International Symposium on Biomedical Imaging (ISBI) in April 2020. The challenge saw a wide participation from across the world, and the top methods were able to match inter-human concordance for the challenge metric. In this paper, we summarize the dataset and the key findings of the challenge, including the commonalities and differences between the methods developed by various participants. We have released the MoNuSAC2020 dataset to the public.

Standard of care diagnostic procedure for suspected skin cancer is microscopic examination of hematoxylin & eosin stained tissue by a pathologist. Areas of high inter-pathologist discordance and rising biopsy rates necessitate higher efficiency and diagnostic reproducibility. We present and validate a deep learning system which classifies digitized dermatopathology slides into 4 categories. The system is developed using 5,070 images from a single lab, and tested on an uncurated set of 13,537 images from 3 test labs, using whole slide scanners manufactured by 3 different vendors. The system’s use of deep-learning-based confidence scoring as a criterion to consider the result as accurate yields an accuracy of up to 98%, and makes it adoptable in a real-world setting. Without confidence scoring, the system achieved an accuracy of 78%. We anticipate that our deep learning system will serve as a foundation enabling faster diagnosis of skin cancer, identification of cases for specialist review, and targeted diagnostic classifications.

The exponential spread of the COVID-19 pandemic has caused countries to impose drastic measures on the public including social distancing, movement restrictions and lockdowns. These government interventions have led to different mobility patterns for the populations. We propose a method of causal inference using community mobility datasets to determine the treatment effects of government interventions on population mobility related outcomes. We first identify the changepoint based on the data of government interventions. We also perform changepoint detection to verify that there is indeed a changepoint at the time of intervention. Then we estimate the mobility trends using a Bayesian structural causal model and project the counterfactual. This is compared to the actual values after interventions to give the treatment effect of interventions. As a specific example, we analyze mobility trends in India before and after interventions. Our analysis shows that there are significant changes in mobility due to government interventions. Our paper aims to provide insights into changes in response to government measures and we hope that it is helpful to those making critical decisions to combat COVID-19.

In this paper, we build autoencoders to learn a latent space from unlabeled image datasets obtained from the Mars rover. Then, once the latent feature space has been learnt, we use k-means to cluster the data. We test the performance of the algorithm on a smaller labeled dataset, and report good accuracy and concordance with the ground truth labels. This is the first attempt to use deep learning based unsupervised algorithms to cluster Mars Rover images. This algorithm can be used to augment human annotations for such datasets (which are time consuming) and speed up the generation of ground truth labels for Mars Rover image data, and potentially other planetary and space images.

In this paper, we estimate the effect of heat stress index (a measure which takes into account rising temperatures as well as humidity) on data center energy consumption. We use forecasting models to predict future energy use by data centers, taking into account rising temperature scenarios. We compare those estimates with baseline forecasted energy consumption (without heat stress index or rising temperature correction) and present the result that there is a sizeable and significant difference in the two forecasts. We show that rising temperatures will cause a negative impact on data center energy consumption, increasing it by about 8 percent, and conclude that data center energy consumption analyses and forecasts must include the effects of heat stress index and rising temperatures and other climate change related effects.

We propose a method for causal inference using satellite image time series, in order to determine the treatment effects of interventions which impact climate change, such as deforestation. Simply put, the aim is to quantify the 'before versus after' effect of climate related human driven interventions, such as urbanization; as well as natural disasters, such as hurricanes and forest fires. As a concrete example, we focus on quantifying forest tree cover change/ deforestation due to human led causes. The proposed method involves the following steps. First, we uae computer vision and machine learning/deep learning techniques to detect and quantify forest tree coverage levels over time, at every time epoch. We then look at this time series to identify changepoints. Next, we estimate the expected (forest tree cover) values using a Bayesian structural causal model and projecting/forecasting the counterfactual. This is compared to the values actually observed post intervention, and the difference in the two values gives us the effect of the intervention (as compared to the non intervention scenario, i.e. what would have possibly happened without the intervention). As a specific use case, we analyze deforestation levels before and after the hyperinflation event (intervention) in Brazil (which ended in 1993-94), for the Amazon rainforest region, around Rondonia, Brazil. For this deforestation use case, using our causal inference framework can help causally attribute change/reduction in forest tree cover and increasing deforestation rates due to human activities at various points in time.

In this paper, we predict severity of extreme weather events (tropical storms, hurricanes, etc.) using buoy data time series variables such as wind speed and air temperature. The prediction/forecasting method is based on various forecasting and machine learning models. The following steps are used. Data sources for the buoys and weather events are identified, aggregated and merged. For missing data imputation, we use Kalman filters as well as splines for multivariate time series. Then, statistical tests are run to ascertain increasing trends in weather event severity. Next, we use machine learning to predict/forecast event severity using buoy variables, and report good accuracies for the models built.

With the recent advances in machine learning and the deep learning paradigm, there is a huge demand to push the data analytics and cognitive inference to the edge of the network near the data producers and sensors. Edge analytics are essential for real-time video analytics and situational awareness; which is required for the wide range of cyber-physical applications such as smart transportation, smart cities, and smart health. To this end, novel architectures and platforms are required to enable real-time low-power deep learning execution at the edge. This paper introduces a novel reconfigurable architecture for real-time execution of deep learning and in particular convolutional Neural Networks (CNNs) at the edge of the network, close to the video camera. The proposed architecture offers a set of coarse-grain function blocks required for realizing CNN algorithms. The macro-pipelined datapath is created by chaining the function blocks with respect to the topology of the target network. The function blocks operate over the streaming pixels (directly fed from the camera interface) in a producer/consumer fashion. At the same time, function blocks offer enough flexibility to adjust the processing with respect to area, power, and performance requirements. This paper primarily focuses on the two first layers of CNNs as the two most compute-intensive layers of CNN network. Our implementation on Xilinx Zynq FPGAs, for the first two layers of the SqueezNet Network, shows 315 mW power consumption when designed at 30 fps, with only a 0.24 ms one-time-latency. In contrast, the Nvidia Tegra TX2 GPU is limited to perform at 32.2 fps due to the 31.4 ms delay, with a much higher power consumption (7.5 W).

We explore a new technique for video frame rate up-conversion. A noniterative multilayer motion estimation algorithm is investigated, based on spatio-temporal smoothness constraints. For regions in the interpolated frame which cannot be motion compensated, we use an exemplar based video inpainting algorithm. The proposed approach yields excellent results compared to other previous approaches

In this paper, we propose the use of causal inference techniques for survival function estimation and prediction for subgroups of the data, upto individual units. Tree ensemble methods, specifically random forests were modified for this purpose. A real world healthcare dataset was used with about 1800 patients with breast cancer, which has multiple patient covariates as well as disease free survival days (DFS) and a death event binary indicator (y). We use the type of cancer curative intervention as the treatment variable (T=0 or 1, binary treatment case in our example). The algorithm is a 2 step approach. In step 1, we estimate heterogeneous treatment effects using a causalTree with the DFS as the dependent variable. Next, in step 2, for each selected leaf of the causalTree with distinctly different average treatment effect (with respect to survival), we fit a survival forest to all the patients in that leaf, one forest each for treatment T=0 as well as T=1 to get estimated patient level survival curves for each treatment (more generally, any model can be used at this step). Then, we subtract the patient level survival curves to get the differential survival curve for a given patient, to compare the survival function as a result of the 2 treatments. The path to a selected leaf also gives us the combination of patient features and their values which are causally important for the treatment effect difference at the leaf.