Lawrence Hall

Abhiram Kandiyana, Ankur Mali, Lawrence O. Hall et al.

Deep-learning pipelines for microscopy image classification often require expensive, labor- and time-intensive expert annotation to produce high-quality ground truth for training. Recent work has shown that prompt tuning of vision-language models (VLMs) can reduce manual annotation by constructing a small prompt set of expert-verified image-caption exemplars that is reused as few-shot context to classify all remaining images at inference time. To further reduce effort, the VLM can draft captions for candidate exemplars, which experts then verify and lightly edit instead of writing text de novo. However, two practical questions remain unaddressed: (1) which unlabeled images should be prioritized for verification, and (2) how many verified exemplars are needed to reach a performance target. In this work, we address these questions by formulating prompt-set construction as a target-driven active learning problem that prioritizes which images to annotate. We study three complementary selection criteria under strict low-resource constraints with small unlabeled pools. Experiments show that our methods reach the target performance with substantially fewer expert-verified images than random selection, achieving 100% test accuracy with as few as 20 annotated images on average. More broadly, our human-in-the-loop framework demonstrates a human-centered use of generative AI in biomedical image analysis, where experts remain actively involved in verifying and refining model output while significantly reducing annotation cost. Code and data will be publicly available.

Duong Mai, Lawrence Hall

Deep learned (DL) models for image recognition have been shown to fail to generalize to data from different devices, populations, etc. COVID-19 detection from Chest X-rays (CXRs), in particular, has been shown to fail to generalize to out-of-distribution (OOD) data from new clinical sources not covered in the training set. This occurs because models learn to exploit shortcuts - source-specific artifacts that do not translate to new distributions - rather than reasonable biomarkers to maximize performance on in-distribution (ID) data. Rendering the models more robust to distribution shifts, our study investigates the use of fundamental noise injection techniques (Gaussian, Speckle, Poisson, and Salt and Pepper) during training. Our empirical results demonstrate that this technique can significantly reduce the performance gap between ID and OOD evaluation from 0.10-0.20 to 0.01-0.06, based on results averaged over ten random seeds across key metrics such as AUC, F1, accuracy, recall and specificity. Our source code is publicly available at https://github.com/Duongmai127/Noisy-ood

Kumar Shashwat, Francis Hahn, Xinming Ou et al.

Large language models (LLM) are perceived to offer promising potentials for automating security tasks, such as those found in security operation centers (SOCs). As a first step towards evaluating this perceived potential, we investigate the use of LLMs in software pentesting, where the main task is to automatically identify software security vulnerabilities in source code. We hypothesize that an LLM-based AI agent can be improved over time for a specific security task as human operators interact with it. Such improvement can be made, as a first step, by engineering prompts fed to the LLM based on the responses produced, to include relevant contexts and structures so that the model provides more accurate results. Such engineering efforts become sustainable if the prompts that are engineered to produce better results on current tasks, also produce better results on future unknown tasks. To examine this hypothesis, we utilize the OWASP Benchmark Project 1.2 which contains 2,740 hand-crafted source code test cases containing various types of vulnerabilities. We divide the test cases into training and testing data, where we engineer the prompts based on the training data (only), and evaluate the final system on the testing data. We compare the AI agent's performance on the testing data against the performance of the agent without the prompt engineering. We also compare the AI agent's results against those from SonarQube, a widely used static code analyzer for security testing. We built and tested multiple versions of the AI agent using different off-the-shelf LLMs -- Google's Gemini-pro, as well as OpenAI's GPT-3.5-Turbo and GPT-4-Turbo (with both chat completion and assistant APIs). The results show that using LLMs is a viable approach to build an AI agent for software pentesting that can improve through repeated use and prompt engineering.

Abhiram Kandiyana, Peter R. Mouton, Yaroslav Kolinko et al.

Traditional deep learning-based methods for classifying cellular features in microscopy images require time- and labor-intensive processes for training models. Among the current limitations are major time commitments from domain experts for accurate ground truth preparation; and the need for a large amount of input image data. We previously proposed a solution that overcomes these challenges using OpenAI's GPT-4(V) model on a pilot dataset (Iba-1 immuno-stained tissue sections from 11 mouse brains). Results on the pilot dataset were equivalent in accuracy and with a substantial improvement in throughput efficiency compared to the baseline using a traditional Convolutional Neural Net (CNN)-based approach. The present study builds upon this framework using a second unique and substantially larger dataset of microscopy images. Our current approach uses a newer and faster model, GPT-4o, along with improved prompts. It was evaluated on a microscopy image dataset captured at low (10x) magnification from cresyl-violet-stained sections through the cerebellum of a total of 18 mouse brains (9 Lurcher mice, 9 wild-type controls). We used our approach to classify these images either as a control group or Lurcher mutant. Using 6 mice in the prompt set the results were correct classification for 11 out of the 12 mice (92%) with 96% higher efficiency, reduced image requirements, and lower demands on time and effort of domain experts compared to the baseline method (snapshot ensemble of CNN models). These results confirm that our approach is effective across multiple datasets from different brain regions and magnifications, with minimal overhead.

Ankur Mali, Lawrence Hall, Jake Williams et al.

Activation functions govern the expressivity and stability of neural networks, yet existing comparisons remain largely heuristic. We propose a rigorous framework for their classification via a nine-dimensional integral signature S_sigma(phi), combining Gaussian propagation statistics (m1, g1, g2, m2, eta), asymptotic slopes (alpha_plus, alpha_minus), and regularity measures (TV(phi'), C(phi)). This taxonomy establishes well-posedness, affine reparameterization laws with bias, and closure under bounded slope variation. Dynamical analysis yields Lyapunov theorems with explicit descent constants and identifies variance stability regions through (m2', g2). From a kernel perspective, we derive dimension-free Hessian bounds and connect smoothness to bounded variation of phi'. Applying the framework, we classify eight standard activations (ReLU, leaky-ReLU, tanh, sigmoid, Swish, GELU, Mish, TeLU), proving sharp distinctions between saturating, linear-growth, and smooth families. Numerical Gauss-Hermite and Monte Carlo validation confirms theoretical predictions. Our framework provides principled design guidance, moving activation choice from trial-and-error to provable stability and kernel conditioning.

Fred Mubang, Lawrence Hall

The Volume-Audience-Match simulator, or VAM was applied to predict future activity on Twitter related to international economic affairs. VAM was applied to do timeseries forecasting to predict the: (1) number of total activities, (2) number of active old users, and (3) number of newly active users over the span of 24 hours from the start time of prediction. VAM then used these volume predictions to perform user link predictions. A user-user edge was assigned to each of the activities in the 24 future timesteps. VAM considerably outperformed a set of baseline models in both the time series and user-assignment tasks

Subhashis Banerjee, Sushmita Mitra, Lawrence O. Hall

Prediction of Overall Survival (OS) of brain cancer patients from multi-modal MRI is a challenging field of research. Most of the existing literature on survival prediction is based on Radiomic features, which does not consider either non-biological factors or the functional neurological status of the patient(s). Besides, the selection of an appropriate cut-off for survival and the presence of censored data create further problems. Application of deep learning models for OS prediction is also limited due to the lack of large annotated publicly available datasets. In this scenario we analyse the potential of two novel neuroimaging feature families, extracted from brain parcellation atlases and spatial habitats, along with classical radiomic and geometric features; to study their combined predictive power for analysing overall survival. A cross validation strategy with grid search is proposed to simultaneously select and evaluate the most predictive feature subset based on its predictive power. A Cox Proportional Hazard (CoxPH) model is employed for univariate feature selection, followed by the prediction of patient-specific survival functions by three multivariate parsimonious models viz. Coxnet, Random survival forests (RSF) and Survival SVM (SSVM). The brain cancer MRI data used for this research was taken from two open-access collections TCGA-GBM and TCGA-LGG available from The Cancer Imaging Archive (TCIA). Corresponding survival data for each patient was downloaded from The Cancer Genome Atlas (TCGA). A high cross validation $C-index$ score of $0.82\pm.10$ was achieved using RSF with the best $24$ selected features. Age was found to be the most important biological predictor. There were $9$, $6$, $6$ and $2$ features selected from the parcellation, habitat, radiomic and region-based feature groups respectively.

Kaoutar Ben Ahmed, Lawrence O. Hall, Dmitry B. Goldgof et al.

Identifying who is infected with the Covid-19 virus is critical for controlling its spread. X-ray machines are widely available worldwide and can quickly provide images that can be used for diagnosis. A number of recent studies claim it may be possible to build highly accurate models, using deep learning, to detect Covid-19 from chest X-ray images. This paper explores the robustness and generalization ability of convolutional neural network models in diagnosing Covid-19 disease from frontal-view (AP/PA), raw chest X-ray images that were lung field cropped. Some concerning observations are made about high performing models that have learned to rely on confounding features related to the data source, rather than the patient's lung pathology, when differentiating between Covid-19 positive and negative labels. Specifically, these models likely made diagnoses based on confounding factors such as patient age or image processing artifacts, rather than medically relevant information.

Lawrence O. Hall, Rahul Paul, Dmitry B. Goldgof et al.

Testing for COVID-19 has been unable to keep up with the demand. Further, the false negative rate is projected to be as high as 30% and test results can take some time to obtain. X-ray machines are widely available and provide images for diagnosis quickly. This paper explores how useful chest X-ray images can be in diagnosing COVID-19 disease. We have obtained 122 chest X-rays of COVID-19 and over 4,000 chest X-rays of viral and bacterial pneumonia. A pretrained deep convolutional neural network has been tuned on 102 COVID-19 cases and 102 other pneumonia cases in a 10-fold cross validation. The results were all 102 COVID-19 cases were correctly classified and there were 8 false positives resulting in an AUC of 0.997. On a test set of 20 unseen COVID-19 cases all were correctly classified and more than 95% of 4171 other pneumonia examples were correctly classified. This study has flaws, most critically a lack of information about where in the disease process the COVID-19 cases were and the small data set size. More COVID-19 case images will enable a better answer to the question of how useful chest X-rays can be for diagnosing COVID-19 (so please send them).

Saeed S. Alahmari, Dmitry Goldgof, Lawrence O. Hall et al.

Lack of enough labeled data is a major problem in building machine learning based models when the manual annotation (labeling) is error-prone, expensive, tedious, and time-consuming. In this paper, we introduce an iterative deep learning based method to improve segmentation and counting of cells based on unbiased stereology applied to regions of interest of extended depth of field (EDF) images. This method uses an existing machine learning algorithm called the adaptive segmentation algorithm (ASA) to generate masks (verified by a user) for EDF images to train deep learning models. Then an iterative deep learning approach is used to feed newly predicted and accepted deep learning masks/images (verified by a user) to the training set of the deep learning model. The error rate in unbiased stereology count of cells on an unseen test set reduced from about 3 % to less than 1 % after 5 iterations of the iterative deep learning based unbiased stereology process.