Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Pathology reports”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Optimal vocabulary selection approaches for privacy-preserving deep NLP model training for information extraction and cancer epidemiology

With the use of artificial intelligence and machine learning techniques for biomedical informatics, security and privacy concerns over the data and subject identities have also become an important issue and essential research topic. Without intentional safeguards, machine learning models may find patterns and features to improve task performance that are associated with private personal information. The privacy vulnerability of deep learning models for information extraction from medical textural contents needs to be quantified since the models are exposed to private health information and personally identifiable information. The objective of the study is to quantify the privacy vulnerability of the deep learning models for natural language processing and explore a proper way of securing patients’ information to mitigate confidentiality breaches. The target model is the multitask convolutional neural network for information extraction from cancer pathology reports, where the data for training the model are from multiple state population-based cancer registries. This study proposes the following schemes to collect vocabularies from the cancer pathology reports; (a) words appearing in multiple registries, and (b) words that have higher mutual information. We performed membership inference attacks on the models in high-performance computing environments. The comparison outcomes suggest that the proposed vocabulary selection methods resulted in lower privacy vulnerability while maintaining the same level of clinical task performance.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Deformable phrase level attention: A flexible approach for improving AI based medical coding

Objective: Improving the AI-driven automated medical encoding of clinical text plays a vital role in gathering information on the occurrence of diseases to improve population-level health. This work presents a novel attention mechanism designed to enhance text classification models and ensure appropriate classification of medical concepts in unstructured electronic health records. Materials and Methods: We developed a deformable, phrase-level attention mechanism to identify important lexical word-level and contextual phrase-level information from clinical text documents. We evaluated conventional and transformer-based deep learning models that we extended with our attention mechanism on the extraction of critical cancer information (e.g., site, subsite, laterality, histology, behavior) from 629,908 electronic pathology reports and on the automated medical encoding of 52,722 hospital discharge summaries. Results: Transformer-based models with the deformable, phrase-level attention mechanism achieved the best performance on the extraction of critical cancer information from pathology reports. Conventional- and transformer-based models show similar or better performance than their baseline counterparts on the automated medical encoding of clinical documents. Discussion: The addition of phrase-level information allowed models extended with our proposed method to outperform standard word-level attention. Our method showed favorable properties for the real-world application in terms of model robustness and phenotyping. These results indicate that our method is promising for automated data harmonization for common data models. Conclusion: This work proposes a novel deformable, phrase-level attention mechanism that enhances text classification models in the extraction of medical concepts from clinical text documents. We demonstrate strong performances on two clinical text datasets and showcase real-world deployability of our method.

Automated medical encoding↗
DOE OSTI2021

Class imbalance in out-of-distribution datasets: Improving the robustness of the TextCNN for the classification of rare cancer types

In the last decade, the widespread adoption of electronic health record documentation has created huge opportunities for information mining. Natural language processing (NLP) techniques using machine and deep learning are becoming increasingly widespread for information extraction tasks from unstructured clinical notes. Disparities in performance when deploying machine learning models in the real world have recently received considerable attention. In the clinical NLP domain, the robustness of convolutional neural networks (CNNs) for classifying cancer pathology reports under natural distribution shifts remains understudied. In this research, we aim to quantify and improve the performance of the CNN for text classification on out-of-distribution (OOD) datasets resulting from the natural evolution of clinical text in pathology reports. We identified class imbalance due to different prevalence of cancer types as one of the sources of performance drop and analyzed the impact of previous methods for addressing class imbalance when deploying models in real-world domains. Our results show that our novel class-specialized ensemble technique outperforms other methods for the classification of rare cancer types in terms of macro F1 scores. We also found that traditional ensemble methods perform better in top classes, leading to higher micro F1 scores. Based on our findings, we formulate a series of recommendations for other ML practitioners on how to build robust models with extremely imbalanced datasets in biomedical NLP applications.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

Multimodal Data Representation with Deep Learning for Extracting Cancer Characteristics from Clinical Text

This paper presents a multimodal data representation to improve the performance of deep learning models for extracting cancer key characteristics from unstructured text in pathology reports. Specifically, in addition to using the text as the input to deep learning models, we use concept unique identifiers (CUIs) as another source of information to the models. We analyze the performance of different text and CUI data representations, including word embeddings and bag of embeddings (BOE), with a convolutional neural network (CNN) and a fully connected multilayer perceptron neural network (MLP-NN). The high level document embeddings from text and CUI inputs are combined by concatenating them and then applying a classifier. The model is used for extracting cancer subsite and histology from pathology reports. These two classification tasks have a large number of labels, i.e. 317 for subsite and 556 for histology, with extreme class imbalance. We compare the performance of the developed DL models across the two tasks based on micro- and macro-F1 scores. The evaluation shows that a multi-channel DL model that utilizes text represented by word embeddings and CUIs represented by BOE outperforms other DL models. Also, this approach significantly improves the model performance on low prevalence classes.

Alawad, Mohammed↗
DOE OSTI2019

Adversarial Training for Privacy-Preserving Deep Learning Model Distribution

Collaboration among cancer registries is essential to develop accurate, robust, and generalizable deep learning models for automated information extraction from cancer pathology reports. Sharing data presents a serious privacy issue, especially in biomedical research and healthcare delivery domains. Distributing pretrained deep learning (DL) models has been proposed to avoid critical data sharing. However, there is growing recognition that collaboration among clinical institutes through DL model distribution exposes new security and privacy vulnerabilities. These vulnerabilities increase in natural language processing (NLP) applications, in which the dataset vocabulary with word vector representations needs to be associated with the other model parameters. In this paper, we propose a novel privacy-preserving DL model distribution across cancer registries for information extraction from cancer pathology reports with privacy and confidentiality considerations. The proposed approach exploits the adversarial training framework to distinguish private features from shared features among different datasets. It only shares registry-invariant model parameters, without sharing raw data nor registry-specific model parameters among cancer registries. Thus, it protects both the data and the trained model simultaneously. We compare our proposed approach to single-registry models, and a model trained on centrally hosted data from different cancer registries. The results show that the proposed approach significantly outperforms the single-registry models and achieves statistically indistinguishable micro and macro F1-score as compared to the centralized model.

Alawad, Mohammed↗
DOE OSTI2016

Fluoroscopy-Guided Percutaneous Vertebral Body Biopsy Using a Novel Drill-Powered Device: Technical Case Series

BackgroundA novel coaxial biopsy system powered by a handheld drill has recently been introduced for percutaneous bone biopsy. This technical note describes our initial experience performing fluoroscopy-guided vertebral body biopsies with this system, compares the yield of drill-assisted biopsy specimens with those obtained using a manual technique, and assesses the histologic adequacy of specimens obtained with drill assistance.MethodsMedical records of all single-level, fluoroscopy-guided vertebral body biopsies were reviewed. Procedural complications were documented according to the Society of Interventional Radiology classification. The total length of bone core obtained from drill-assisted biopsies was compared with that of matched manual biopsies. Pathology reports were reviewed to determine the histologic adequacy of specimens obtained with drill assistance.ResultsTwenty eight drill-assisted percutaneous vertebral body biopsies met study inclusion criteria. No acute complications were reported. Of the 86 % (24/28) of patients with clinical follow-up, no delayed complications were reported (median follow-up, 28 weeks; range 5–115 weeks). The median total length of bone core obtained from drill-assisted biopsies was 28 mm (range 8–120 mm). This was longer than that obtained from manual biopsies (median, 20 mm; range 5–45 mm; P = 0.03). Crush artifact was present in 11 % (3/28) of drill-assisted biopsy specimens, which in one case (3.6 %; 1/28) precluded definitive diagnosis.ConclusionsA drill-assisted, coaxial biopsy system can be used to safely obtain vertebral body core specimens under fluoroscopic guidance. The higher bone core yield obtained with drill assistance may be offset by the presence of crush artifact.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
NASA NTRS2016

Colonoscopy Screening in the US Astronaut Corps

BACKGROUND: Historically, colonoscopy screenings for astronauts have been conducted to ensure that astronauts are in good health for space missions. Recently this historical data has been identified as being useful for developing an occupational surveillance requirement. It can be used to assess overall colon health and to have a point of reference for future tests in current and former astronauts, as well as to follow-up and track rates of colorectal cancer and polyps. These rates can be compared to military and other terrestrial populations. In 2003, the active astronaut colonoscopy requirements changed to require less frequent colonoscopies. Since polyp removal during a colonoscopy is an intervention that prevents the polyp from potentially developing into cancer, the procedure decreases the individual's risk for colon cancer. The objective of this study is to evaluate the possible effect of increased follow-up times between colonoscopies on the number and severity of polyps identified during the procedures among both current and former NASA astronauts. Initial results and forward work regarding astronaut colonoscopy screenings will be presented. METHODS: A retrospective study of all colonoscopy procedures performed on NASA astronauts between 1962 and 2015 (both during active career and retirement) was conducted by review of the JSC Clinic Electronic Medical Record and Lifetime Surveillance of Astronaut Health (LSAH) database for colonoscopy screening procedures and pathology reports. The timeframe of interest was from the time of selection into the Astronaut Corps through May 2015 or death. For each colonoscopy report, the following data were captured: date of procedure, age at time of procedure, reason for procedure, quality of bowel prep, completion of procedure and/or reason for termination of procedure, findings of procedure, subsequent treatment (if any), recommended follow-up interval, actual follow up interval, family history of polyps or colon cancer, and other significant items or discrepancies. The population consisted of 338 astronauts: 52 females, 286 males. Of these, 56 were deceased, and 11 astronauts had no record of any colonoscopies. Because of a screening requirement change in 2003, analyses were conducted to determine if there were differences between the two time periods. One-sided Wilcoxon rank sum tests were used to identify statistically significant differences between the two time periods. RESULTS: There was a combined total of 1,964 colonoscopy screenings identified. The average follow-up intervals between colonoscopies were indeed longer after the screening requirement change than before the change. The mean follow-up interval pre -2003 was 3.59 years, while post-2003 it increased to 4.35 years. The statistical significance of this difference was confirmed using a one sided Wilcoxon rank sum test which yielded p<.001. Colonoscopies performed after the requirement change tended to have a higher incidence and greater severity of polyps. From pre-2003 to post-2003 the percentage of colonoscopy procedures yielding no polyps decreased from 83.77% to 74.70%. Not only did post-2003 procedures yield more polyp findings, but the polyps recorded were more often of severe pathology. Before 2003 3.62% of colonoscopy findings were polyps of the hyperplastic type (the least severe polyp type) and only 3.35% were of greater severity. Post-2003, 4.21% of findings were hyperplastic polyps while 11.44% were of greater severity. Upon the investigation of other possible contributing factors to these results, we also found that mean age post-2003 was 54.55 years which was significantly higher than during the pre-2003 timeframe (47.32 years). This was observed with a one-sided Wilcoxon rank sum test, resulting in a p<0.001. The increased average age of astronauts could also be a contributing factor to the greater number of polyps found since the risk of developing polyps increases with age. Further work is needed to better understand the increased incidence and greater severity of polyps found in astronaut colonoscopy outcomes.

Masterova, K.↗
DOE OSTI2021

Privacy-Preserving Knowledge Transfer with Bootstrap Aggregation of Teacher Ensembles

There is a need to transfer knowledge among institutions and organizations to save effort in annotation and labeling or in enhancing task performance. However, knowledge transfer is difficult because of restrictions that are in place to ensure data security and privacy. Institutions are not allowed to exchange data or perform any activity that may expose personal information. With the leverage of a differential privacy algorithm in a high-performance computing environment, we propose a new training protocol, Bootstrap Aggregation of Teacher Ensembles (BATE), which is applicable to various types of machine learning models. The BATE algorithm is based on and provides enhancements to the PATE algorithm, maintaining competitive task performance scores on complex datasets with underrepresented class labels.We conducted a proof-of-the-concept study of the information extraction from cancer pathology report data from four cancer registries and performed comparisons between four scenarios: no collaboration, no privacy-preserving collaboration, the PATE algorithm, and the proposed BATE algorithm. The results showed that the BATE algorithm maintained competitive macro-averaged F1 scores, demonstrating that the suggested algorithm is an effective yet privacy-preserving method for machine learning and deep learning solutions.

Yoon, Hong-Jun↗
DOE OSTI2019

Can magnetic resonance imaging differentiate among transurethral bulking agent, urethral diverticulum, and periurethral cyst?

Purpose: To evaluate magnetic resonance imaging findings that differentiate among periurethral bulking agents (primarily collagen), urethral diverticulum, and periurethral cyst. Methods: We searched our radiologic database retrospectively from 2001 to 2017 for periurethral cystic lesions, identifying a total of 50 patients with 68 lesions. Final diagnoses in 68 lesions were bulking agents (27), urethral diverticula (29), and periurethral cysts (12). Two abdominal radiologists, blinded to clinical history, independently evaluated T1, T2, and post-contrast images. The readers assessed number, morphological features, location, connection to urethra and mass effect, signal intensity, and enhancement for each lesion. Fisher exact test and logistic regression analysis were performed for each univariate significant feature. The operative and pathologic reports were the reference standard. Results: Magnetic resonance imaging features found more often in bulking agents versus urethral diverticulum were multiple lesions (P = 0.011), upper or upper-mid-urethral location (P ≤ 0.0001), lack of internal fluid/fluid level (P = 0.002), no urethral connection (P = 0.005), T1 isointensity, and T2 mild hyperintensity compared to muscles but lower T2 signal than urine (P < 0.0001). Most cases of urethral diverticula and periurethral cysts were detected at mid- and lower urethra. Urethral diverticula were larger than bulking agents and periurethral cysts (P = 0.005 and P = 0.023) (mean diameter = 24, 16, 15 mm, respectively). Most bulking agents (93%) and urethral diverticula (90%) showed mass effect on urethra, while periurethral cysts (75%) did not (P < 0.0001). Conclusion: Signal intensity and lesion characterization on magnetic resonance imaging can significantly differentiate bulking agent from urethral diverticulum and periurethral cyst. Radiologists should consider differential diagnosis of a bulking agent, especially when distinguishing characteristics described here are present to prevent incorrect diagnosis and ultimately unnecessary surgical intervention.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2017

Germline Chd8 haploinsufficiency alters brain development in mouse

The chromatin remodeling gene CHD8 represents a central node in neurodevelopmental gene networks implicated in autism. In this paper, we examined the impact of germline heterozygous frameshift Chd8 mutation on neurodevelopment in mice. Chd8 + / del5 mice displayed normal social interactions with no repetitive behaviors but exhibited cognitive impairment correlated with increased regional brain volume, validating that phenotypes of Chd8 + / del5 mice overlap pathology reported in humans with CHD8 mutations. We applied network analysis to characterize neurodevelopmental gene expression, revealing widespread transcriptional changes in Chd8 + / del5 mice across pathways disrupted in neurodevelopmental disorders, including neurogenesis, synaptic processes and neuroimmune signaling. We identified a co-expression module with peak expression in early brain development featuring dysregulation of RNA processing, chromatin remodeling and cell-cycle genes enriched for promoter binding by Chd8, and we validated increased neuronal proliferation and developmental splicing perturbation in Chd8 + / del5 mice. Finally, this integrative analysis offers an initial picture of the consequences of Chd8 haploinsufficiency for brain development.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Pitfalls in the n -mode representation of vibrational potentials

Simulations of anharmonic vibrational motion rely on computationally expedient representations of the governing potential energy surface. The n-mode representation (n-MR)—effectively a many-body expansion in the space of molecular vibrations—is a general and efficient approach that is often used for this purpose in vibrational self-consistent field (VSCF) calculations and correlated analogues thereof. In the present analysis, a lack of convergence in many VSCF calculations is shown to originate from negative and unbound potentials at truncated orders of the n-MR expansion. For cases of strong anharmonic coupling between modes, the n-MR can both dip below the true global minimum of the potential surface and lead to effective single-mode potentials in VSCF that do not correspond to bound vibrational problems, even for bound total potentials. The present analysis serves mainly as a pathology report of this issue. Furthermore, this insight into the origin of VSCF non-convergence provides a simple, albeit ad hoc, route to correct the problem by “painting in” the full representation of groups of modes that exhibit these negative potentials at little additional computational cost. Somewhat surprisingly, this approach also reasonably approximates the results of the next-higher n-MR order and identifies groups of modes with particularly strong coupling. The method is shown to identify and correct problematic triples of modes—and restore SCF convergence—in two-mode representations of challenging test systems, including the water dimer and trimer, as well as protonated tropine.

Chemistry↗
DOE OSTI2025

Multi-Label Classification with Constraint-Based Learning for Hierarchical Consistency

We explore the limitations of traditional crossentropy loss in a hierarchical multi-label classification setting and introduce a novel loss function. This function is designed to integrate hierarchical constraints directly into the training process. By incorporating such constraints into the loss, our approach slightly improves the logical consistency of predictions in structured domains. We demonstrate the efficacy of our approach through experiments on primary site and histology classification by using electronic pathology reports. These results show that our proposed hierarchical loss function enhances the model's ability to produce predictions that are logically consistent with the natural data hierarchies, and it slightly improves predictive accuracy. Our framework may be extended to other hierarchical domains, however the performance gains are context specific.

Shivanna, Abhishek [ORNL] (ORCID:0009000665228593)↗
DOE OSTI2023

Characterizing Quantum Classifier Utility in Natural Language Processing Workflows

Quantum Natural Language Processing (QNLP) develops natural language processing (NLP) models for deployment on quantum computers. We explore feature and data prototype selection techniques to address challenges posed by encoding high dimensional features. Our study builds quantum circuit classifiers that includes classical feature pre-processing, quantum embedding and quantum model training. The quantum models are built on 4 or 6 qubits and the quantum neural network (QNN) uses the established bricklayer design. We compare the dependence of model performance (in terms of accuracy and F1 scores) on feature length, embedding gates and parameterized unitary design. We compare the performance of quantum machine learning models to classical convolution neural network model (CNN) on binary and multi-class classification tasks using two datasets of synthetic features and labels. The first is the ECP-CANDLE P3B3 dataset a corpus of synthetically generated cancer pathology reports. The second dataset is extracted from well-known benchmark dataset (MADELON) - features are generated with a combination of informative, repeated and uninformative features. Both datasets are used for binary classification and multi-class classification with 3 classes. We observe robust, accurate performance from all models on the binary classification tasks, but multiclass classification is a challenge for the quantum models-there is a notable decrease in accuracy when using 3 classes. Overall the performance is comparable in terms of recall and accuracy between QNNs and CNNs, even with large datasets. These results provide a point of comparison between quantum and classical models on real-world datasets.

Hamilton, Kathleen↗
DOE OSTI2018

Scalable deep text comprehension for Cancer surveillance on high-performance computing

Background: Deep Learning (DL) has advanced the state-of-the-art capabilities in bioinformatics applications which has resulted in trends of increasingly sophisticated and computationally demanding models trained by larger and larger data sets. This vastly increased computational demand challenges the feasibility of conducting cutting-edge research. One solution is to distribute the vast computational workload across multiple computing cluster nodes with data parallelism algorithms. In this study, we used a High-Performance Computing environment and implemented the Downpour Stochastic Gradient Descent algorithm for data parallelism to train a Convolutional Neural Network (CNN) for the natural language processing task of information extraction from a massive dataset of cancer pathology reports. We evaluated the scalability improvements using data parallelism training and the Titan supercomputer at Oak Ridge Leadership Computing Facility. To evaluate scalability, we used different numbers of worker nodes and performed a set of experiments comparing the effects of different training batch sizes and optimizer functions.Results: We found that Adadelta would consistently converge at a lower validation loss, though requiring over twice as many training epochs as the fastest converging optimizer, RMSProp. The Adam optimizer consistently achieved a close 2nd place minimum validation loss significantly faster; using a batch size of 16 and 32 allowed the network to converge in only 4.5 training epochs.Conclusions: We demonstrated that the networked training process is scalable across multiple compute nodes communicating with message passing interface while achieving higher classification accuracy compared to a traditional machine learning algorithm.

Qiu, John X.↗
DOE OSTI2026

Mitigating Algorithmic Bias in Cancer Site Classification Models

Purpose Integrating artificial intelligence in cancer diagnostics has improved tumor classification beyond rule-based systems. Despite these advancements, these models may still encode demographic biases. We conducted a large-scale, applied bias-probing study of a deep learning–based cancer site classifier to quantify race information encoded in document embeddings. We then evaluated how performance changes when race-correlated embedding dimensions are removed in a post-training sensitivity analysis. Methods The cancer site classifier was trained using 3.5 million electronic cancer pathology reports from six of the National Cancer Institute's SEER registries. We trained a hierarchical self-attention network to generate 400-dimensional document embeddings. These embeddings were used to train two downstream, gradient-boosted decision tree classifiers: one to classify the cancer sites and another to predict racial categories. We identified overlapping features by intersecting the top 50 feature-importance rankings from the site and race models and computed their cumulative feature importance in each model. As a post hoc sensitivity analysis, we progressively pruned these overlapping dimensions, retrained the site model, and compared overall macro-F1 and accuracy, race-stratified macro-F1, and group fairness metrics on the basis of demographic parity and equalized odds before and after pruning. Results The analysis revealed minimal feature overlap between the cancer site and race prediction models, and the cumulative importance scores indicated a negligible influence of racial information on clinical predictions. Post-training pruning of overlapping features did not compromise the models' diagnostic accuracy, with a 0.07% loss in accuracy. Conclusion Our findings demonstrate that HiSAN-generated embeddings from SEER data can be used effectively in cancer site classification without significant demographic bias influencing the outcomes. Post-training pruning therefore functions as a practical audit and sensitivity check.

Shivanna, Abhishek [ORNL] (ORCID:0009000665228593)↗
DOE OSTI2025

Risks Associated with Sharing the MOSSAIC APIs

The MOSSAIC APIs contain two files which, in theory, could be used to discover information about the pathology report data from the SEER registries on which the AI models were trained. In this document, we explain the contents of these files and assess the associated risk. APPENDIX A contains a set of slides to aid in the dissemination of this information.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1968

Color-televised medical microscopy

Color television microscopy used at laboratory range magnifications, reproduces a slide image with sufficient fidelity for medical laboratory and instructional use. The system is used for instant pathological reporting between operating room and remotely located pathologist viewing a biopsy through this medium.

Heath, M. A.↗
DOE OSTI2018

CAT: computer aided triage improving upon the Bayes risk through ε-refusal triage rules

Manual extraction of information from electronic pathology (epath) reports to populate the Surveillance, Epidemiology, and End Result (SEER) database is labor intensive. Systematizing the data extraction automatically using machine-learning (ML) and natural language processing (NLP) is desirable to reduce the human labor required to populate the SEER database and to improve the timeliness of the data. This enables scaling up registry efficiency and collection of new data elements. To ensure the integrity, quality, and continuity of the SEER data, the misclassification error of ML and NPL algorithms needs to be negligible. Current algorithms fail to achieve the precision of human experts who can bring additional information in their assessments. Differences in registry format and the desire to develop a common information extraction platform further complicate the ML/NLP tasks. The purpose of our study is to develop triage rules to partially automate registry workflow to improve the precision of the auto-extracted information. This paper presents a mathematical framework to improve the precision of a classifier beyond that of the Bayes classifier by selectively classifying item that are most likely to be correct. This results in a triage rule that only classifies a subset of the item. We characterize the optimal triage rule and demonstrate its usefulness in the problem of classifying cancer site from electronic pathology reports to achieve a desired precision. From the mathematical formalism, we propose a heuristic estimate for triage rule based on post-processing the soft-max output from standard machine learning algorithms. We show, in test cases, that the triage rule significantly improve the classification accuracy.

60 APPLIED LIFE SCIENCES↗