Conversational Chatbot for Cigarette Smoking Cessation: Report of the User-Centered Design Eleven Step Development Process.

BACKGROUND: Conversational chatbots are an emerging digital intervention for smoking cessation. No studies have reported on the entire development process of a cessation chatbot. OBJECTIVE: To describe the user-centered design development process for a novel and comprehensive quit smoking conversational chatbot called "QuitBot." METHODS: The four years of formative research for developing QuitBot followed an eleven-step process: (1) specifying a conceptual model, (2) conducting content analysis of existing interventions (63 hours of intervention transcripts), (3) assessing user needs, (4) developing the chat's persona ("personality"), (5) prototyping content and persona, (6) developing full functionality, (7) programming the QuitBot, (8) conducting a diary study, (9) conducting a pilot randomized trial, (10) reviewing results of the trial, and (11) adding a free-form question and answer (QnA) function, based on user feedback from pilot trial results. The process of adding a QnA function itself involved a three-step process: (a) generating QnA pairs, (b) fine tuning Large Language Models (LLMs) on QnA pairs, and (c) evaluating the LLM model outputs. RESULTS: A quit smoking program spanning 42 days of 2 to 3-minute conversations covering topics ranging from motivations to quit, setting a quit date, choosing FDA-approved cessation medications, coping with triggers, and recovering from lapses/relapses. In a pilot randomized trial with 96% three-month outcome data retention, QuitBot demonstrated high user engagement and promising cessation rates compared to the National Cancer Institute's SmokefreeTXT (SFT) text messaging program-particularly among those who viewed all 42 days of program content: 30-day complete-case, point prevalence abstinence (PPA) rates at three-month follow-up were 63% (39/62) for QuitBot vs. 38% (45/117) for SFT (OR = 2.58; 95% CI: 1.34, 4.99; P =.005). However, Facebook Messenger (FM) intermittently blocked participants' access to QuitBot so we transitioned from FM to a standalone smartphone app as the communication channel. Participants' frustration with QuitBot's inability to answer their open-ended questions lead to us develop a core conversational feature enabling users to ask open-ended questions about quitting cigarette smoking and for the QuitBot to respond with accurate and professional answers. To support this functionality, we developed a library of 11,000 QnA pairs on topics associated with quitting cigarette smoking. Model testing results showed that Microsoft's Azure-based QnA maker effectively handled questions that matched our library of 11,000 QnA pairs. A fine-tuned, contextualized GPT3.5 responds to questions that are not within our library of QnA pairs. CONCLUSIONS: The development process yielded the first LLM-based quit smoking program delivered as a conversational chatbot. Iterative testing led to significant enhancements, including improvements to the delivery channel. A pivotal addition was the inclusion of a core LLM-supported conversational feature allowing users to ask open-ended questions. CLINICALTRIAL: ClinicalTrials.gov Identifier, NCT03585231.

Democratizing protein language models with parameter-efficient fine-tuning.

Proteomics has been revolutionized by large protein language models (PLMs), which learn unsupervised representations from large corpora of sequences. These models are typically fine-tuned in a supervised setting to adapt the model to specific downstream tasks. However, the computational and memory footprint of fine-tuning (FT) large PLMs presents a barrier for many research groups with limited computational resources. Natural language processing has seen a similar explosion in the size of models, where these challenges have been addressed by methods for parameter-efficient fine-tuning (PEFT). In this work, we introduce this paradigm to proteomics through leveraging the parameter-efficient method LoRA and training new models for two important tasks: predicting protein-protein interactions (PPIs) and predicting the symmetry of homooligomer quaternary structures. We show that these approaches are competitive with traditional FT while requiring reduced memory and substantially fewer parameters. We additionally show that for the PPI prediction task, training only the classification head also remains competitive with full FT, using five orders of magnitude fewer parameters, and that each of these methods outperform state-of-the-art PPI prediction methods with substantially reduced compute. We further perform a comprehensive evaluation of the hyperparameter space, demonstrate that PEFT of PLMs is robust to variations in these hyperparameters, and elucidate where best practices for PEFT in proteomics differ from those in natural language processing. All our model adaptation and evaluation code is available open-source at https://github.com/microsoft/peft_proteomics. Thus, we provide a blueprint to democratize the power of PLM adaptation to groups with limited computational resources.

Improved estimation of the relationship between fetal growth and late stillbirth in the United States, 2014-15.

In the United States the rate of stillbirth after 28 weeks' gestation (late stillbirth) is 2.7/1000 births. Fetuses that are small for gestational age (SGA) or large for gestational age (LGA) are at increased risk of stillbirth. SGA and LGA are often categorized as growth or birthweight ≤ 10th and ≥ 90th centile, respectively; however, these cut-offs are arbitrary. We sought to characterize the relationship between birthweight and stillbirth risk in greater detail. Data on singleton births between 28- and 44-weeks' gestation from 2014 to 2015 were extracted from the US Centers for Disease Control and Prevention live birth and fetal death files. Growth was assessed using customized birthweight centiles (Gestation Related Optimal Weight; GROW). The analyses included logistic regression using SGA/LGA categories and a generalized additive model (GAM) using birthweight centile as a continuous exposure. Although the SGA and LGA categories identified infants at risk of stillbirth, categorical models provided poor fits to the data within the high-risk bins, and in particular markedly underestimated the risk for the extreme centiles. For example, for fetuses in the lowest GROW centile, the observed rate was 39.8/1000 births compared with a predicted rate of 11.7/1000 from the category-based analysis. In contrast, the model-predicted risk from the GAM tracked closely with the observed risk, with the GAM providing an accurate characterization of stillbirth risk across the entire birthweight continuum. This study provides stillbirth risk estimates for each GROW centile, which clinicians can use in conjunction with other clinical details to guide obstetric management.

Assessment of differentially private synthetic data for utility and fairness in end-to-end machine learning pipelines for tabular data.

Differentially private (DP) synthetic datasets are a solution for sharing data while preserving the privacy of individual data providers. Understanding the effects of utilizing DP synthetic data in end-to-end machine learning pipelines impacts areas such as health care and humanitarian action, where data is scarce and regulated by restrictive privacy laws. In this work, we investigate the extent to which synthetic data can replace real, tabular data in machine learning pipelines and identify the most effective synthetic data generation techniques for training and evaluating machine learning models. We systematically investigate the impacts of differentially private synthetic data on downstream classification tasks from the point of view of utility as well as fairness. Our analysis is comprehensive and includes representatives of the two main types of synthetic data generation algorithms: marginal-based and GAN-based. To the best of our knowledge, our work is the first that: (i) proposes a training and evaluation framework that does not assume that real data is available for testing the utility and fairness of machine learning models trained on synthetic data; (ii) presents the most extensive analysis of synthetic dataset generation algorithms in terms of utility and fairness when used for training machine learning models; and (iii) encompasses several different definitions of fairness. Our findings demonstrate that marginal-based synthetic data generators surpass GAN-based ones regarding model training utility for tabular data. Indeed, we show that models trained using data generated by marginal-based algorithms can exhibit similar utility to models trained using real data. Our analysis also reveals that the marginal-based synthetic data generated using AIM and MWEM PGM algorithms can train models that simultaneously achieve utility and fairness characteristics close to those obtained by models trained with real data.

Self-Supervised Learning for Improved Optical Coherence Tomography Detection of Macular Telangiectasia Type 2.

Importance: Deep learning image analysis often depends on large, labeled datasets, which are difficult to obtain for rare diseases. Objective: To develop a self-supervised approach for automated classification of macular telangiectasia type 2 (MacTel) on optical coherence tomography (OCT) with limited labeled data. Design, Setting, and Participants: This was a retrospective comparative study. OCT images from May 2014 to May 2019 were collected by the Lowy Medical Research Institute, La Jolla, California, and the University of Washington, Seattle, from January 2016 to October 2022. Clinical diagnoses of patients with and without MacTel were confirmed by retina specialists. Data were analyzed from January to September 2023. Exposures: Two convolutional neural networks were pretrained using the Bootstrap Your Own Latent algorithm on unlabeled training data and fine-tuned with labeled training data to predict MacTel (self-supervised method). ResNet18 and ResNet50 models were also trained using all labeled data (supervised method). Main Outcomes and Measures: The ground truth yes vs no MacTel diagnosis is determined by retinal specialists based on spectral-domain OCT. The models' predictions were compared against human graders using accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), area under precision recall curve (AUPRC), and area under the receiver operating characteristic curve (AUROC). Uniform manifold approximation and projection was performed for dimension reduction and GradCAM visualizations for supervised and self-supervised methods. Results: A total of 2636 OCT scans from 780 patients with MacTel and 131 patients without MacTel were included from the MacTel Project (mean [SD] age, 60.8 [11.7] years; 63.8% female), and another 2564 from 1769 patients without MacTel from the University of Washington (mean [SD] age, 61.2 [18.1] years; 53.4% female). The self-supervised approach fine-tuned on 100% of the labeled training data with ResNet50 as the feature extractor performed the best, achieving an AUPRC of 0.971 (95% CI, 0.969-0.972), an AUROC of 0.970 (95% CI, 0.970-0.973), accuracy of 0.898%, sensitivity of 0.898, specificity of 0.949, PPV of 0.935, and NPV of 0.919. With only 419 OCT volumes (185 MacTel patients in 10% of labeled training dataset), the ResNet18 self-supervised model achieved comparable performance, with an AUPRC of 0.958 (95% CI, 0.957-0.960), an AUROC of 0.966 (95% CI, 0.964-0.967), and accuracy, sensitivity, specificity, PPV, and NPV of 90.2%, 0.884, 0.916, 0.896, and 0.906, respectively. The self-supervised models showed better agreement with the more experienced human expert graders. Conclusions and Relevance: The findings suggest that self-supervised learning may improve the accuracy of automated MacTel vs non-MacTel binary classification on OCT with limited labeled training data, and these approaches may be applicable to other rare diseases, although further research is warranted.

Known pathogenic gene variants and new candidates detected in Sudden Unexpected Infant Death using Whole Genome Sequencing.

Purpose: To gain insights into potential genetic factors contributing to the infant's vulnerability to Sudden Unexpected Infant Death (SUID). Methods: Whole Genome Sequencing (WGS) was performed on 145 infants that succumbed to SUID, and 576 healthy adults. Variants were filtered by gnomAD allele frequencies and predictions of functional consequences. Results: Variants of interest were identified in 86 genes, 63.4% of our cohort. Seventy-one of these have been previously associated with SIDS/SUID/SUDP. Forty-three can be characterized as cardiac genes and are related to cardiomyopathies, arrhythmias, and other conditions. Variants in 22 genes were associated with neurologic functions. Variants were also found in 13 genes reported to be pathogenic for various systemic disorders. Variants in eight genes are implicated in the response to hypoxia and the regulation of reactive oxygen species (ROS) and have not been previously described in SIDS/SUID/SUDP. Seventy-two infants met the triple risk hypothesis criteria (Figure 1). Conclusion: Our study confirms and further expands the list of genetic variants associated with SUID. The abundance of genes associated with heart disease and the discovery of variants associated with the redox metabolism have important mechanistic implications for the pathophysiology of SUID.

Risk factors for late (28+ weeks' gestation) stillbirth in the United States, 2014-2015.

BACKGROUND: In the United States (US) late stillbirth (at 28 weeks or more of gestation) occurs in 3/1000 births. AIM: We examined risk factors for late stillbirth with the specific goal of identifying modifiable factors that contribute substantially to stillbirth burden. SETTING: All singleton births in the US for 2014-2015. METHODS: We used a retrospective population-based design to assess the effects of multiple factors on the risk of late stillbirth in the US. Data were drawn from the US Centers for Disease Control and Prevention live birth and fetal death data files. RESULTS: There were 6,732,157 live and 18,334 stillbirths available for analysis (late stillbirth rate = 2.72/1000 births). The importance of sociodemographic determinants was shown by higher risks for Black and Native Hawaiian and Other Pacific Islander mothers compared with White mothers, mothers with low educational attainment, and older mothers. Among modifiable risk factors, delayed/absent prenatal care, diabetes, hypertension, and maternal smoking were associated with increased risk, though they accounted for only 3-6% of stillbirths each. Two factors accounted for the largest proportion of late stillbirths: high maternal body mass index (BMI; 15%) and infants who were small for gestational age (38%). Participation in the supplemental nutrition for women, infants and children program was associated with a 28% reduction in overall stillbirth burden. CONCLUSIONS: This study provides population-based evidence for stillbirth risk in the US. A high proportion of late stillbirths was associated with high maternal BMI and small for gestational age, whereas participation in supplemental nutrition programs was associated with a large reduction in stillbirth burden. Addressing obesity and fetal growth restriction, as well as broadening participation in nutritional supplementation programs could reduce late stillbirths.

Automatic segmentation of prostate cancer metastases in PSMA PET/CT images using deep neural networks with weighted batch-wise dice loss.

PURPOSE: Automatic and accurate segmentation of lesions in images of metastatic castration-resistant prostate cancer has the potential to enable personalized radiopharmaceutical therapy and advanced treatment response monitoring. The aim of this study is to develop a convolutional neural networks-based framework for fully-automated detection and segmentation of metastatic prostate cancer lesions in whole-body PET/CT images. METHODS: 525 whole-body PET/CT images of patients with metastatic prostate cancer were available for the study, acquired with the [18F]DCFPyL radiotracer that targets prostate-specific membrane antigen (PSMA). U-Net (1)-based convolutional neural networks (CNNs) were trained to identify lesions on paired axial PET/CT slices. Baseline models were trained using batch-wise dice loss, as well as the proposed weighted batch-wise dice loss (wDice), and the lesion detection performance was quantified, with a particular emphasis on lesion size, intensity, and location. We used 418 images for model training, 30 for model validation, and 77 for model testing. In addition, we allowed our model to take n = 0,2, …, 12 neighboring axial slices to examine how incorporating greater amounts of 3D context influences model performance. We selected the optimal number of neighboring axial slices that maximized the detection rate on the 30 validation images, and trained five neural networks with different architectures. RESULTS: Model performance was evaluated using the detection rate, Dice similarity coefficient (DSC) and sensitivity. We found that the proposed wDice loss significantly improved the lesion detection rate, lesion-wise DSC and lesion-wise sensitivity compared to the baseline, with corresponding average increases of 0.07 (p-value = 0.01), 0.03 (p-value = 0.01) and 0.04 (p-value = 0.01), respectively. The inclusion of the first two neighboring axial slices in the input likewise increased the detection rate by 0.17, lesion-wise DSC by 0.05, and lesion-wise mean sensitivity by 0.16. However, there was a minimal effect from including more distant neighboring slices. We ultimately chose to use a number of neighboring slices equal to 2 and the wDice loss function to train our final model. To evaluate the model's performance, we trained three models using identical hyperparameters on three different data splits. The results showed that, on average, the model was able to detect 80% of all testing lesions, with a detection rate of 93% for lesions with maximum standardized uptake values (SUVmax) greater than 5.0. In addition, the average median lesion-wise DSC was 0.51 and 0.60 for all the lesions and lesions with SUVmax>5.0, respectively, on the testing set. Four additional neural networks with different architectures were trained, and they both yielded stronger performance of segmenting lesions whose SUVmax>5.0 compared to the rest of lesions. CONCLUSION: Our results demonstrate that prostate cancer metastases in PSMA PET/CT images can be detected and segmented using CNNs. The segmentation performance strongly depends on the intensity, size, and the location of lesions, and can be improved by using specialized loss functions. Specifically, the models performed best in detection of lesions with SUVmax>5.0. Another challenge was to accurately segment lesions close to the bladder. Future work will focus on improving the detection of lesions with lower SUV values by designing custom loss functions that take into account the lesion intensity, using additional data augmentation techniques, and reducing the number of false lesions by developing methods to better separate signal from noise.

Maternal infections in pregnancy and the risk of sudden unexpected infant death in the offspring in the U.S., 2011-2015.

BACKGROUND: Infection is thought to play a part in some infant deaths. Maternal infection in pregnancy has focused on chlamydia with some reports suggesting an association with sudden unexpected infant death (SUID). OBJECTIVES: We hypothesized that maternal infections in pregnancy are associated with subsequent SUID in their offspring. SETTING: All births in the United States, 2011-2015. DATA SOURCE: Centers for Disease Control and Prevention (CDC) Birth Cohort Linked Birth-Infant Death Data Files. STUDY DESIGN: Cohort study, although the data were analysed as a case control study. Cases were infants that died from SUID. Controls were randomly sampled infants that survived their first year of life; approximately 10 controls per SUID case. EXPOSURES: Chlamydia, gonorrhea and hepatitis C. RESULTS: There were 19,849,690 live births in the U.S. for the period 2011-2015. There were 37,143 infant deaths of which 17,398 were classified as SUID cases (a rate of 0.86/1000 live births). The proportion of the control mothers with chlamydia was 1.7%, gonorrhea 0.2% and hepatitis C was 0.3%. Chlamydia was present in 3.8% of mothers whose infants subsequently died of SUID compared with 1.7% of controls (unadjusted OR = 2.35, 95% CI = 2.15, 2.56; adjusted OR = 1.08, 95% CI = 0.98, 1.19). Gonorrhea was present in 0.7% of mothers of SUID cases compared with 0.2% of mothers of controls (OR = 3.09, (2.50, 3.79); aOR = 1.20(0.95, 1.49)) and hepatitis C was present in 1.3% of mothers of SUID cases compared with 0.3% of mothers of controls (OR = 4.69 (3.97, 5.52): aOR = 1.80 (1.50, 2.15)). CONCLUSIONS: The marked attenuation of SUID risk after adjustment for a wide variety of socioeconomic and demographic factors suggests the small increase in the risk of SUID of the offspring of mothers with infection with hepatitis C in pregnancy is due to residual confounding.

Factors Associated With the Intention to Receive the COVID-19 Vaccine: Cross-sectional National Study.

BACKGROUND: The COVID-19 pandemic is an unprecedented public health crisis, and vaccines are the most effective means of preventing severe consequences of this disease. Hesitancy regarding vaccines persists among adults in the United States, despite overwhelming scientific evidence of safety and efficacy. OBJECTIVE: The purpose of this study was to use the Health Belief Model (HBM) and reasoned action approach (RAA) to examine COVID-19 vaccine hesitancy by comparing those who had already received 1 vaccine to those who had received none. METHODS: This study examined demographic and theory-based factors associated with vaccine uptake and intention among 1643 adults in the United States who completed an online survey during February and March 2021. Survey items included demographic variables (eg, age, sex, political ideology), attitudes, and health belief variables (eg, perceived self-efficacy, perceived susceptibility). Hierarchical logistic regression analyses were used for vaccine uptake/intent. The first model included demographic variables. The second model added theory-based factors to examine the association of health beliefs and vaccine uptake above and beyond the associations explained by demographic characteristics alone. RESULTS: The majority of participants were male (n=974, 59.3%), White (n=1347, 82.0%), and non-Hispanic (n=1518, 92.4%) and reported they had already received a COVID-19 vaccine or definitely would when it was available to them (n=1306, 79.5%). Demographic variables significantly associated with vaccine uptake/intent included age (adjusted odds ratio [AOR] 1.05, 95% CI 1.04-1.06), other race (AOR 0.47, 95% CI 0.27-0.83 vs White), and political ideology (AOR 15.77, 95% CI 7.03-35.35 very liberal vs very conservative). The theory-based factors most strongly associated with uptake/intention were attitudes (AOR 3.72, 95% CI 2.42-5.73), self-efficacy (AOR 1.75, 95% CI 1.34-2.29), and concerns about side effects (AOR 0.59, 95% CI 0.46-0.76). Although race and political ideology were significant in the model of demographic characteristics, they were not significant when controlling for attitudes and beliefs. CONCLUSIONS: Vaccination represents one of the best tools to combat the COVID-19 pandemic, as well as other possible pandemics in the future. This study showed that older age, attitudes, injunctive norms, descriptive norms, and self-efficacy are positively associated with vaccine uptake and intent, whereas perceived side effects and lack of trust in the vaccine are associated with lower uptake and intent. Race and political ideology were not significant predictors when attitudes and beliefs were considered. Before vaccine hesitancy can be addressed, researchers and clinicians must understand the basis of vaccine hesitancy and which populations may show higher hesitancy to the vaccination so that interventions can be adequately targeted.

Deep learning models for COVID-19 chest x-ray classification: Preventing shortcut learning using feature disentanglement.

In response to the COVID-19 global pandemic, recent research has proposed creating deep learning based models that use chest radiographs (CXRs) in a variety of clinical tasks to help manage the crisis. However, the size of existing datasets of CXRs from COVID-19+ patients are relatively small, and researchers often pool CXR data from multiple sources, for example, using different x-ray machines in various patient populations under different clinical scenarios. Deep learning models trained on such datasets have been shown to overfit to erroneous features instead of learning pulmonary characteristics in a phenomenon known as shortcut learning. We propose adding feature disentanglement to the training process. This technique forces the models to identify pulmonary features from the images and penalizes them for learning features that can discriminate between the original datasets that the images come from. We find that models trained in this way indeed have better generalization performance on unseen data; in the best case we found that it improved AUC by 0.13 on held out data. We further find that this outperforms masking out non-lung parts of the CXRs and performing histogram equalization, both of which are recently proposed methods for removing biases in CXR datasets.

Association of Public Health Measures During the COVID-19 Pandemic With the Incidence of Infectious Conjunctivitis.

Importance: Infectious conjunctivitis is highly transmissible and a public health concern. While mitigation strategies have been successful on a local level, population-wide decreases in spread are rare. Objective: To evaluate whether internet search interest and emergency department visits for infectious conjunctivitis were associated with public health interventions adopted during the COVID-19 pandemic. Design, Setting, and Participants: Internet search data from the US and emergency department data from a single academic center in the US were used in this study. Publicly available smartphone mobility data were temporally aligned to quantify social distancing. Internet search term trends for nonallergic conjunctivitis, corneal abrasions, and posterior vitreous detachments were obtained. Additionally, all patients who presented to a single emergency department from February 2015 to February 2021 were included in a review. Physician notes for emergency department visits at a single academic center with the same diagnoses were extracted. Causal inference was performed using a bayesian structural time-series model. Data were compared from before and after April 2020, when the US Centers for Disease Control and Prevention recommended members of the public wear masks, stay at least 6 feet from others who did not reside in the same home, avoid crowds, and quarantine if experiencing flulike symptoms or exposure to persons with COVID-19 symptoms. Exposures: Symptoms of or interest in conjunctivitis in the context of the COVID-19 pandemic. Main Outcome and Measures: The hypothesis was that there would be a decrease in internet search interest and emergency department visits for infectious conjunctivitis after the adaptation of public health measures targeted to curb COVID-19. Results: A total of 1156 emergency department encounters with a diagnosis of conjunctivitis were noted from January 2015 to February 2021. Emergency department encounters for nonallergic conjunctivitis decreased by 37.3% (95% CI, -12.9% to -60.6%; P < .001). In contrast, encounters for corneal abrasion (1.1% [95% CI, -29.3% to 29.1%]; P = .47) and posterior vitreous detachments (7.9% [95% CI, -46.9% to 66.6%]; P = .39) remained stable after adjusting for total emergency department encounters. Search interest in conjunctivitis decreased by 34.2% (95% CI, -30.6% to -37.6%; P < .001) after widespread implementation of public health interventions to mitigate COVID-19. Conclusions and Relevance: Public health interventions, such as social distancing, increased emphasis on hygiene, and travel restrictions during the COVID-19 pandemic, were associated with decreased search interest in nonallergic conjunctivitis and conjunctivitis-associated emergency department encounters. Mobility data may provide novel metrics of social distancing. These data provide evidence of a sustained population-wide decrease in infectious conjunctivitis.

Detecting, classifying, and counting blue whale calls with Siamese neural networks.

The goal of this project is to use acoustic signatures to detect, classify, and count the calls of four acoustic populations of blue whales so that, ultimately, the conservation status of each population can be better assessed. We used manual annotations from 350 h of audio recordings from the underwater hydrophones in the Indian Ocean to build a deep learning model to detect, classify, and count the calls from four acoustic song types. The method we used was Siamese neural networks (SNN), a class of neural network architectures that are used to find the similarity of the inputs by comparing their feature vectors, finding that they outperformed the more widely used convolutional neural networks (CNN). Specifically, the SNN outperform a CNN with 2% accuracy improvement in population classification and 1.7%-6.4% accuracy improvement in call count estimation for each blue whale population. In addition, even though we treat the call count estimation problem as a classification task and encode the number of calls in each spectrogram as a categorical variable, SNN surprisingly learned the ordinal relationship among them. SNN are robust and are shown here to be an effective way to automatically mine large acoustic datasets for blue whale calls.

Precision mapping child undernutrition for nearly 600,000 inhabited census villages in India.

There are emerging opportunities to assess health indicators at truly small areas with increasing availability of data geocoded to micro geographic units and advanced modeling techniques. The utility of such fine-grained data can be fully leveraged if linked to local governance units that are accountable for implementation of programs and interventions. We used data from the 2011 Indian Census for village-level demographic and amenities features and the 2016 Indian Demographic and Health Survey in a bias-corrected semisupervised regression framework to predict child anthropometric failures for all villages in India. Of the total geographic variation in predicted child anthropometric failure estimates, 54.2 to 72.3% were attributed to the village level followed by 20.6 to 39.5% to the state level. The mean predicted stunting was 37.9% (SD: 10.1%; IQR: 31.2 to 44.7%), and substantial variation was found across villages ranging from less than 5% for 691 villages to over 70% in 453 villages. Estimates at the village level can potentially shift the paradigm of policy discussion in India by enabling more informed prioritization and precise targeting. The proposed methodology can be adapted and applied to diverse population health indicators, and in other contexts, to reveal spatial heterogeneity at a finer geographic scale and identify local areas with the greatest needs and with direct implications for actions to take place.