1
|
Crombé A, Lecomte JC, Seux M, Banaste N, Gorincour G. Using the Textual Content of Radiological Reports to Detect Emerging Diseases: A Proof-of-Concept Study of COVID-19. JOURNAL OF IMAGING INFORMATICS IN MEDICINE 2024; 37:620-632. [PMID: 38343242 PMCID: PMC11031522 DOI: 10.1007/s10278-023-00949-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 08/15/2023] [Revised: 10/02/2023] [Accepted: 10/04/2023] [Indexed: 04/20/2024]
Abstract
Changes in the content of radiological reports at population level could detect emerging diseases. Herein, we developed a method to quantify similarities in consecutive temporal groupings of radiological reports using natural language processing, and we investigated whether appearance of dissimilarities between consecutive periods correlated with the beginning of the COVID-19 pandemic in France. CT reports from 67,368 consecutive adults across 62 emergency departments throughout France between October 2019 and March 2020 were collected. Reports were vectorized using time frequency-inverse document frequency (TF-IDF) analysis on one-grams. For each successive 2-week period, we performed unsupervised clustering of the reports based on TF-IDF values and partition-around-medoids. Next, we assessed the similarities between this clustering and a clustering from two weeks before according to the average adjusted Rand index (AARI). Statistical analyses included (1) cross-correlation functions (CCFs) with the number of positive SARS-CoV-2 tests and advanced sanitary index for flu syndromes (ASI-flu, from open-source dataset), and (2) linear regressions of time series at different lags to understand the variations of AARI over time. Overall, 13,235 chest CT reports were analyzed. AARI was correlated with ASI-flu at lag = + 1, + 5, and + 6 weeks (P = 0.0454, 0.0121, and 0.0042, respectively) and with SARS-CoV-2 positive tests at lag = - 1 and 0 week (P = 0.0057 and 0.0001, respectively). In the best fit, AARI correlated with the ASI-flu with a lag of 2 weeks (P = 0.0026), SARS-CoV-2-positive tests in the same week (P < 0.0001) and their interaction (P < 0.0001) (adjusted R2 = 0.921). Thus, our method enables the automatic monitoring of changes in radiological reports and could help capturing disease emergence.
Collapse
Affiliation(s)
- Amandine Crombé
- IMADIS, Lyon, France.
- SARCOTARGET Team, University of Bordeaux, Inserm, UMR1312, BRIC, BoRdeaux Institute of Oncology, 146 Rue Léo Saignat, Bordeaux, F-33076, France.
- Department of Radiology, Pellegrin University Hospital, CHU Bordeaux, Place Amélie Raba-Léon, Bordeaux, F-33076, France.
| | - Jean-Christophe Lecomte
- IMADIS, Lyon, France
- Centre Aquitain d'Imagerie médicale, Mérignac, France
- Centre Hospitalier de Saintes, Saintes, France
- Clinique Mutualiste Bordeaux Pessac, Pessac, France
| | | | - Nathan Banaste
- IMADIS, Lyon, France
- Clinique Convert, Ramsay, Bourg en Bresse, France
| | | |
Collapse
|
2
|
Melnyk O, Ismail A, Ghorashi NS, Heekin M, Javan R. Generative Artificial Intelligence Terminology: A Primer for Clinicians and Medical Researchers. Cureus 2023; 15:e49890. [PMID: 38174178 PMCID: PMC10762565 DOI: 10.7759/cureus.49890] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 12/04/2023] [Indexed: 01/05/2024] Open
Abstract
Generative artificial intelligence (AI) is rapidly transforming the medical field, as advanced tools powered by large language models (LLMs) make their way into clinical practice, research, and education. Chatbots, which can generate human-like responses, have gained attention for their potential applications. Therefore, familiarity with LLMs and other promising generative AI tools is crucial to harness their potential safely and effectively. As these AI-based technologies continue to evolve, medical professionals must develop a strong understanding of AI terminologies and concepts, particularly generative AI, to effectively tackle real-world challenges and create solutions. This knowledge will enable healthcare professionals to utilize AI-driven innovations for improved patient care and increased productivity in the future. In this brief technical report, we explore 20 of the most relevant terminology associated with the underlying technology behind LLMs and generative AI as they relate to the medical field and provide some examples of how these topics relate to healthcare applications to help in their understanding.
Collapse
Affiliation(s)
- Oleksiy Melnyk
- Department of Radiology, George Washington University School of Medicine and Health Sciences, Washington D.C., USA
| | - Ahmed Ismail
- Department of Radiology, George Washington University School of Medicine and Health Sciences, Washington D.C., USA
| | - Nima S Ghorashi
- Department of Radiology, George Washington University School of Medicine and Health Sciences, Washington D.C., USA
| | - Mary Heekin
- Department of Radiology, George Washington University School of Medicine and Health Sciences, Washington D.C., USA
| | - Ramin Javan
- Department of Radiology, George Washington University School of Medicine and Health Sciences, Washington D.C., USA
| |
Collapse
|
3
|
Nugen F, Vera Garcia DV, Sohn S, Mickley JP, Wyles CC, Erickson BJ, Taunton MJ. Application of Natural Language Processing in Total Joint Arthroplasty: Opportunities and Challenges. J Arthroplasty 2023; 38:1948-1953. [PMID: 37619802 DOI: 10.1016/j.arth.2023.08.047] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/14/2022] [Revised: 08/10/2023] [Accepted: 08/11/2023] [Indexed: 08/26/2023] Open
Abstract
Total joint arthroplasty is becoming one of the most common surgeries within the United States, creating an abundance of analyzable data to improve patient experience and outcomes. Unfortunately, a large majority of this data is concealed in electronic health records only accessible by manual extraction, which takes extensive time and resources. Natural language processing (NLP), a field within artificial intelligence, may offer a viable alternative to manual extraction. Using NLP, a researcher can analyze written and spoken data and extract data in an organized manner suitable for future research and clinical use. This article will first discuss common subtasks involved in an NLP pipeline, including data preparation, modeling, analysis, and external validation, followed by examples of NLP projects. Challenges and limitations of NLP will be discussed, closing with future directions of NLP projects, including large language models.
Collapse
Affiliation(s)
- Fred Nugen
- Orthopedic Surgery Artificial Intelligence Laboratory (OSAIL), Mayo Clinic, Rochester, Minnesota; Radiology Informatics Lab (RIL), Department of Radiology, Mayo Clinic, Rochester, Minnesota
| | - Diana V Vera Garcia
- Orthopedic Surgery Artificial Intelligence Laboratory (OSAIL), Mayo Clinic, Rochester, Minnesota; Radiology Informatics Lab (RIL), Department of Radiology, Mayo Clinic, Rochester, Minnesota
| | - Sunghwan Sohn
- Orthopedic Surgery Artificial Intelligence Laboratory (OSAIL), Mayo Clinic, Rochester, Minnesota; Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota
| | - John P Mickley
- Orthopedic Surgery Artificial Intelligence Laboratory (OSAIL), Mayo Clinic, Rochester, Minnesota; Department of Orthopedic Surgery, Mayo Clinic, Rochester, Minnesota
| | - Cody C Wyles
- Orthopedic Surgery Artificial Intelligence Laboratory (OSAIL), Mayo Clinic, Rochester, Minnesota; Department of Orthopedic Surgery, Mayo Clinic, Rochester, Minnesota
| | - Bradley J Erickson
- Radiology Informatics Lab (RIL), Department of Radiology, Mayo Clinic, Rochester, Minnesota
| | - Michael J Taunton
- Orthopedic Surgery Artificial Intelligence Laboratory (OSAIL), Mayo Clinic, Rochester, Minnesota; Department of Orthopedic Surgery, Mayo Clinic, Rochester, Minnesota
| |
Collapse
|
4
|
Wibaek R, Andersen GS, Dahm CC, Witte DR, Hulman A. Large Language Models for Epidemiological Research via Automated Machine Learning: Case Study Using Data From the British National Child Development Study. JMIR Med Inform 2023; 11:e43638. [PMID: 37787655 PMCID: PMC10547934 DOI: 10.2196/43638] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/21/2022] [Revised: 06/29/2023] [Accepted: 07/22/2023] [Indexed: 10/04/2023] Open
Abstract
Background Large language models have had a huge impact on natural language processing (NLP) in recent years. However, their application in epidemiological research is still limited to the analysis of electronic health records and social media data. objectives To demonstrate the potential of NLP beyond these domains, we aimed to develop prediction models based on texts collected from an epidemiological cohort and compare their performance to classical regression methods. Methods We used data from the British National Child Development Study, where 10,567 children aged 11 years wrote essays about how they imagined themselves as 25-year-olds. Overall, 15% of the data set was set aside as a test set for performance evaluation. Pretrained language models were fine-tuned using AutoTrain (Hugging Face) to predict current reading comprehension score (range: 0-35) and future BMI and physical activity (active vs inactive) at the age of 33 years. We then compared their predictive performance (accuracy or discrimination) with linear and logistic regression models, including demographic and lifestyle factors of the parents and children from birth to the age of 11 years as predictors. Results NLP clearly outperformed linear regression when predicting reading comprehension scores (root mean square error: 3.89, 95% CI 3.74-4.05 for NLP vs 4.14, 95% CI 3.98-4.30 and 5.41, 95% CI 5.23-5.58 for regression models with and without general ability score as a predictor, respectively). Predictive performance for physical activity was similarly poor for the 2 methods (area under the receiver operating characteristic curve: 0.55, 95% CI 0.52-0.60 for both) but was slightly better than random assignment, whereas linear regression clearly outperformed the NLP approach when predicting BMI (root mean square error: 4.38, 95% CI 4.02-4.74 for NLP vs 3.85, 95% CI 3.54-4.16 for regression). The NLP approach did not perform better than simply assigning the mean BMI from the training set as a predictor. Conclusions Our study demonstrated the potential of using large language models on text collected from epidemiological studies. The performance of the approach appeared to depend on how directly the topic of the text was related to the outcome. Open-ended questions specifically designed to capture certain health concepts and lived experiences in combination with NLP methods should receive more attention in future epidemiological studies.
Collapse
Affiliation(s)
| | | | | | - Daniel R Witte
- Department of Public Health, Aarhus University, Aarhus, Denmark
- Steno Diabetes Center Aarhus, Aarhus University Hospital, Aarhus, Denmark
| | - Adam Hulman
- Department of Public Health, Aarhus University, Aarhus, Denmark
- Steno Diabetes Center Aarhus, Aarhus University Hospital, Aarhus, Denmark
| |
Collapse
|
5
|
Basilio R, Carvalho AR, Rodrigues R, Conrado M, Accorsi S, Forghani R, Machuca T, Zanon M, Altmayer S, Hochhegger B. Natural Language Processing for the Identification of Incidental Lung Nodules in Computed Tomography Reports: A Quality Control Tool. JCO Glob Oncol 2023; 9:e2300191. [PMID: 37769221 PMCID: PMC10581645 DOI: 10.1200/go.23.00191] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/12/2023] [Revised: 07/09/2023] [Accepted: 08/22/2023] [Indexed: 09/30/2023] Open
Abstract
PURPOSE To evaluate the diagnostic performance of a natural language processing (NLP) model in detecting incidental lung nodules (ILNs) in unstructured chest computed tomography (CT) reports. METHODS All unstructured consecutive reports of chest CT scans performed at a tertiary hospital between 2020 and 2021 were retrospectively reviewed (n = 21,542) to train the NLP tool. Internal validation was performed using reference readings by two radiologists of both CT scans and reports, using a different external cohort of 300 chest CT scans. Second, external validation was performed in a cohort of all random unstructured chest CT reports from 57 different hospitals conducted in May 2022. A review by the same thoracic radiologists was used as the gold standard. The sensitivity, specificity, and accuracy were calculated. RESULTS Of 21,542 CT reports, 484 mentioned at least one ILN (mean age, 71 ± 17.6 [standard deviation] years; women, 52%) and were included in the training set. In the internal validation (n = 300), the NLP tool detected ILN with a sensitivity of 100.0% (95% CI, 97.6 to 100.0), a specificity of 95.9% (95% CI, 91.3 to 98.5), and an accuracy of 98.0% (95% CI, 95.7 to 99.3). In the external validation (n = 977), the NLP tool yielded a sensitivity of 98.4% (95% CI, 94.5 to 99.8), a specificity of 98.6% (95% CI, 97.5 to 99.3), and an accuracy of 98.6% (95% CI, 97.6 to 99.2). Twelve months after the initial reports, 8 (8.60%) patients had a final diagnosis of lung cancer, among which 2 (2.15%) would have been lost to follow-up without the NLP tool. CONCLUSION NLP can be used to identify ILNs in unstructured reports with high accuracy, allowing a timely recall of patients and a potential diagnosis of early-stage lung cancer that might have been lost to follow-up.
Collapse
Affiliation(s)
- Rodrigo Basilio
- D'Or Institute for Research and Education (IDOR), Rio de Janeiro, Brazil
| | | | - Rosana Rodrigues
- D'Or Institute for Research and Education (IDOR), Rio de Janeiro, Brazil
| | - Marco Conrado
- D'Or Institute for Research and Education (IDOR), Rio de Janeiro, Brazil
| | - Sephania Accorsi
- D'Or Institute for Research and Education (IDOR), Rio de Janeiro, Brazil
| | - Reza Forghani
- Radiomics and Augmented Intelligence Laboratory (RAIL), University of Florida, Gainesville, FL
| | - Tiago Machuca
- D'Or Institute for Research and Education (IDOR), Rio de Janeiro, Brazil
| | - Matheus Zanon
- Federal University of Health Sciences of Porto Alegre, Porto Alegre, Brazil
| | - Stephan Altmayer
- Stanford Hospital, Stanford University Medical Center, Palo Alto, CA
| | - Bruno Hochhegger
- Radiomics and Augmented Intelligence Laboratory (RAIL), University of Florida, Gainesville, FL
- Federal University of Health Sciences of Porto Alegre, Porto Alegre, Brazil
| |
Collapse
|
6
|
Galbusera F, Cina A, Bassani T, Panico M, Sconfienza LM. Automatic Diagnosis of Spinal Disorders on Radiographic Images: Leveraging Existing Unstructured Datasets With Natural Language Processing. Global Spine J 2023; 13:1257-1266. [PMID: 34219477 PMCID: PMC10416592 DOI: 10.1177/21925682211026910] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/16/2022] Open
Abstract
STUDY DESIGN Retrospective study. OBJECTIVES Huge amounts of images and medical reports are being generated in radiology departments. While these datasets can potentially be employed to train artificial intelligence tools to detect findings on radiological images, the unstructured nature of the reports limits the accessibility of information. In this study, we tested if natural language processing (NLP) can be useful to generate training data for deep learning models analyzing planar radiographs of the lumbar spine. METHODS NLP classifiers based on the Bidirectional Encoder Representations from Transformers (BERT) model able to extract structured information from radiological reports were developed and used to generate annotations for a large set of radiographic images of the lumbar spine (N = 10 287). Deep learning (ResNet-18) models aimed at detecting radiological findings directly from the images were then trained and tested on a set of 204 human-annotated images. RESULTS The NLP models had accuracies between 0.88 and 0.98 and specificities between 0.84 and 0.99; 7 out of 12 radiological findings had sensitivity >0.90. The ResNet-18 models showed performances dependent on the specific radiological findings with sensitivities and specificities between 0.53 and 0.93. CONCLUSIONS NLP generates valuable data to train deep learning models able to detect radiological findings in spine images. Despite the noisy nature of reports and NLP predictions, this approach effectively mitigates the difficulties associated with the manual annotation of large quantities of data and opens the way to the era of big data for artificial intelligence in musculoskeletal radiology.
Collapse
Affiliation(s)
| | - Andrea Cina
- IRCCS Istituto Ortopedico Galeazzi, Milan, Italy
| | - Tito Bassani
- IRCCS Istituto Ortopedico Galeazzi, Milan, Italy
| | - Matteo Panico
- IRCCS Istituto Ortopedico Galeazzi, Milan, Italy
- Department of Chemistry, Materials and Chemical Engineering “Giulio Natta,” Politecnico di Milano, Milan, Italy
| | - Luca Maria Sconfienza
- IRCCS Istituto Ortopedico Galeazzi, Milan, Italy
- Department of Biomedical Sciences for Health, Università degli Studi di Milano, Milan, Italy
| |
Collapse
|
7
|
Bikdeli B, Lo YC, Khairani CD, Bejjani A, Jimenez D, Barco S, Mahajan S, Caraballo C, Secemsky EA, Klok FA, Hunsaker AR, Aghayev A, Muriel A, Wang Y, Hussain MA, Appah-Sampong A, Lu Y, Lin Z, Aneja S, Khera R, Goldhaber SZ, Zhou L, Monreal M, Krumholz HM, Piazza G. Developing Validated Tools to Identify Pulmonary Embolism in Electronic Databases: Rationale and Design of the PE-EHR+ Study. Thromb Haemost 2023; 123:649-662. [PMID: 36809777 PMCID: PMC11200175 DOI: 10.1055/a-2039-3222] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/23/2023]
Abstract
BACKGROUND Contemporary pulmonary embolism (PE) research, in many cases, relies on data from electronic health records (EHRs) and administrative databases that use International Classification of Diseases (ICD) codes. Natural language processing (NLP) tools can be used for automated chart review and patient identification. However, there remains uncertainty with the validity of ICD-10 codes or NLP algorithms for patient identification. METHODS The PE-EHR+ study has been designed to validate ICD-10 codes as Principal Discharge Diagnosis, or Secondary Discharge Diagnoses, as well as NLP tools set out in prior studies to identify patients with PE within EHRs. Manual chart review by two independent abstractors by predefined criteria will be the reference standard. Sensitivity, specificity, and positive and negative predictive values will be determined. We will assess the discriminatory function of code subgroups for intermediate- and high-risk PE. In addition, accuracy of NLP algorithms to identify PE from radiology reports will be assessed. RESULTS A total of 1,734 patients from the Mass General Brigham health system have been identified. These include 578 with ICD-10 Principal Discharge Diagnosis codes for PE, 578 with codes in the secondary position, and 578 without PE codes during the index hospitalization. Patients within each group were selected randomly from the entire pool of patients at the Mass General Brigham health system. A smaller subset of patients will also be identified from the Yale-New Haven Health System. Data validation and analyses will be forthcoming. CONCLUSIONS The PE-EHR+ study will help validate efficient tools for identification of patients with PE in EHRs, improving the reliability of efficient observational studies or randomized trials of patients with PE using electronic databases.
Collapse
Affiliation(s)
- Behnood Bikdeli
- Division of Cardiovascular Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
- Thrombosis Research Group, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
- YNHH/Yale Center for Outcomes Research and Evaluation (CORE), New Haven, Connecticut, United States
- Cardiovascular Research Foundation (CRF), New York, New York, United States
| | - Ying-Chih Lo
- Division of General Internal Medicine and Primary Care, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
| | - Candrika D Khairani
- Thrombosis Research Group, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
| | - Antoine Bejjani
- Thrombosis Research Group, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
| | - David Jimenez
- Respiratory Department, Hospital Ramón y Cajal and Medicine Department, Universidad de Alcalá (Instituto de Ramón y Cajal de Investigación Sanitaria), Centro de Investigación Biomédica en Red de Enfermedades Respiratorias, Madrid, Spain
| | - Stefano Barco
- Department of Angiology, University Hospital Zurich, Zurich, Switzerland
- Center for Thrombosis and Hemostasis, Johannes Gutenberg University Mainz, Mainz, Germany
| | - Shiwani Mahajan
- YNHH/Yale Center for Outcomes Research and Evaluation (CORE), New Haven, Connecticut, United States
- Department of Internal Medicine, Yale School of Medicine, New Haven, Connecticut, United States
| | - César Caraballo
- YNHH/Yale Center for Outcomes Research and Evaluation (CORE), New Haven, Connecticut, United States
| | - Eric A Secemsky
- Richard A. and Susan F. Smith Center for Outcomes Research in Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts, United States
- Harvard Medical School, Boston, Massachusetts, United States
- Division of Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts, United States
| | - Frederikus A Klok
- Department of Medicine - Thrombosis and Hemostasis, Leiden University Medical Centre, Leiden, The Netherlands
| | - Andetta R Hunsaker
- Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
| | - Ayaz Aghayev
- Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
| | - Alfonso Muriel
- Clinical Biostatistics Unit. Hospital Universitario Ramón y Cajal. IRYCIS, CIBERESP: Universidad de Alcalá. Madrid, Spain
| | - Yun Wang
- YNHH/Yale Center for Outcomes Research and Evaluation (CORE), New Haven, Connecticut, United States
- Richard A. and Susan F. Smith Center for Outcomes Research in Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts, United States
| | - Mohamad A Hussain
- Division of Vascular and Endovascular Surgery, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
- Centre for Surgery and Public Health, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
| | - Abena Appah-Sampong
- Department of Surgery, Brigham and Women's Hospital, Boston, Massachusetts, United States
| | - Yuan Lu
- YNHH/Yale Center for Outcomes Research and Evaluation (CORE), New Haven, Connecticut, United States
| | - Zhenqiu Lin
- YNHH/Yale Center for Outcomes Research and Evaluation (CORE), New Haven, Connecticut, United States
| | - Sanjay Aneja
- Department of Therapeutic Radiology, Yale School of Medicine, New Haven, Connecticut, United States
| | - Rohan Khera
- YNHH/Yale Center for Outcomes Research and Evaluation (CORE), New Haven, Connecticut, United States
- Section of Cardiovascular Medicine, Department of Internal Medicine, Yale School of Medicine, New Haven, Connecticut, United States
| | - Samuel Z Goldhaber
- Division of Cardiovascular Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
- Thrombosis Research Group, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
| | - Li Zhou
- Division of General Internal Medicine and Primary Care, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
| | - Manuel Monreal
- Cátedra de Enfermedad Tromboembólica, Universidad Católica de Murcia, Murcia, Spain
| | - Harlan M Krumholz
- YNHH/Yale Center for Outcomes Research and Evaluation (CORE), New Haven, Connecticut, United States
- Section of Cardiovascular Medicine, Department of Internal Medicine, Yale School of Medicine, New Haven, Connecticut, United States
- Department of Health Policy and Management, Yale School of Public Health, New Haven, Connecticut, United States
| | - Gregory Piazza
- Division of Cardiovascular Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
- Thrombosis Research Group, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States
| |
Collapse
|
8
|
Albahra S, Gorbett T, Robertson S, D'Aleo G, Kumar SVS, Ockunzzi S, Lallo D, Hu B, Rashidi HH. Artificial intelligence and machine learning overview in pathology & laboratory medicine: A general review of data preprocessing and basic supervised concepts. Semin Diagn Pathol 2023; 40:71-87. [PMID: 36870825 DOI: 10.1053/j.semdp.2023.02.002] [Citation(s) in RCA: 15] [Impact Index Per Article: 15.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/27/2023] [Revised: 02/10/2023] [Accepted: 02/14/2023] [Indexed: 02/17/2023]
Abstract
Machine learning (ML) is becoming an integral aspect of several domains in medicine. Yet, most pathologists and laboratory professionals remain unfamiliar with such tools and are unprepared for their inevitable integration. To bridge this knowledge gap, we present an overview of key elements within this emerging data science discipline. First, we will cover general, well-established concepts within ML, such as data type concepts, data preprocessing methods, and ML study design. We will describe common supervised and unsupervised learning algorithms and their associated common machine learning terms (provided within a comprehensive glossary of terms that are discussed within this review). Overall, this review will offer a broad overview of the key concepts and algorithms in machine learning, with a focus on pathology and laboratory medicine. The objective is to provide an updated useful reference for those new to this field or those who require a refresher.
Collapse
Affiliation(s)
- Samer Albahra
- Pathology and Laboratory Medicine Institute (PLMI), Cleveland Clinic, Cleveland, OH, United States; PLMI's Center for Artificial Intelligence & Data Science, Cleveland Clinic, Cleveland, OH, United States.
| | - Tom Gorbett
- Pathology and Laboratory Medicine Institute (PLMI), Cleveland Clinic, Cleveland, OH, United States; PLMI's Center for Artificial Intelligence & Data Science, Cleveland Clinic, Cleveland, OH, United States
| | - Scott Robertson
- Pathology and Laboratory Medicine Institute (PLMI), Cleveland Clinic, Cleveland, OH, United States; PLMI's Center for Artificial Intelligence & Data Science, Cleveland Clinic, Cleveland, OH, United States
| | - Giana D'Aleo
- Pathology and Laboratory Medicine Institute (PLMI), Cleveland Clinic, Cleveland, OH, United States; PLMI's Center for Artificial Intelligence & Data Science, Cleveland Clinic, Cleveland, OH, United States
| | - Sushasree Vasudevan Suseel Kumar
- Pathology and Laboratory Medicine Institute (PLMI), Cleveland Clinic, Cleveland, OH, United States; PLMI's Center for Artificial Intelligence & Data Science, Cleveland Clinic, Cleveland, OH, United States
| | - Samuel Ockunzzi
- Pathology and Laboratory Medicine Institute (PLMI), Cleveland Clinic, Cleveland, OH, United States; PLMI's Center for Artificial Intelligence & Data Science, Cleveland Clinic, Cleveland, OH, United States
| | - Daniel Lallo
- Pathology and Laboratory Medicine Institute (PLMI), Cleveland Clinic, Cleveland, OH, United States; PLMI's Center for Artificial Intelligence & Data Science, Cleveland Clinic, Cleveland, OH, United States
| | - Bo Hu
- Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH, United States; PLMI's Center for Artificial Intelligence & Data Science, Cleveland Clinic, Cleveland, OH, United States
| | - Hooman H Rashidi
- Pathology and Laboratory Medicine Institute (PLMI), Cleveland Clinic, Cleveland, OH, United States; PLMI's Center for Artificial Intelligence & Data Science, Cleveland Clinic, Cleveland, OH, United States.
| |
Collapse
|
9
|
Meerwijk EL, Tamang SR, Finlay AK, Ilgen MA, Reeves RM, Harris AHS. Suicide theory-guided natural language processing of clinical progress notes to improve prediction of veteran suicide risk: protocol for a mixed-method study. BMJ Open 2022; 12:e065088. [PMID: 36002210 PMCID: PMC9413184 DOI: 10.1136/bmjopen-2022-065088] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 05/24/2022] [Accepted: 08/02/2022] [Indexed: 11/17/2022] Open
Abstract
INTRODUCTION The state-of-the-art 3-step Theory of Suicide (3ST) describes why people consider suicide and who will act on their suicidal thoughts and attempt suicide. The central concepts of 3ST-psychological pain, hopelessness, connectedness, and capacity for suicide-are among the most important drivers of suicidal behaviour but they are missing from clinical suicide risk prediction models in use at the US Veterans Health Administration (VHA). These four concepts are not systematically recorded in structured fields of VHA's electronic healthcare records. Therefore, this study will develop a domain-specific ontology that will enable automated extraction of these concepts from clinical progress notes using natural language processing (NLP), and test whether NLP-based predictors for these concepts improve accuracy of existing VHA suicide risk prediction models. METHODS AND ANALYSIS Our mixed-method study has an exploratory sequential design where a qualitative component (aim 1) will inform quantitative analyses (aims 2 and 3). For aim 1, subject matter experts will manually annotate progress notes of clinical encounters with veterans who attempted or died by suicide to develop a domain-specific ontology for the 3ST concepts. During aim 2, we will use NLP to machine-annotate clinical progress notes and derive longitudinal representations for each patient with respect to the presence and intensity of hopelessness, psychological pain, connectedness and capacity for suicide in temporal proximity of suicide attempts and deaths by suicide. These longitudinal representations will be evaluated during aim 3 for their ability to improve existing VHA prediction models of suicide and suicide attempts, STORM (Stratification Tool for Opioid Risk Mitigation) and REACHVET (Recovery Engagement and Coordination for Health - Veterans Enhanced Treatment). ETHICS AND DISSEMINATION Ethics approval for this study was granted by the Stanford University Institutional Review Board and the Research and Development Committee of the VA Palo Alto Health Care System. Results of the study will be disseminated through several outlets, including peer-reviewed publications and presentations at national conferences.
Collapse
Affiliation(s)
- Esther Lydia Meerwijk
- VA Health Services Research & Development, Center for Innovation to Implementation, VA Palo Alto Health Care System, Palo Alto, California, USA
| | - Suzanne R Tamang
- VA Health Services Research & Development, Center for Innovation to Implementation, VA Palo Alto Health Care System, Palo Alto, California, USA
- Department of Biomedical Data Science, Stanford University, Stanford, California, USA
| | - Andrea K Finlay
- VA Health Services Research & Development, Center for Innovation to Implementation, VA Palo Alto Health Care System, Palo Alto, California, USA
- Schar School of Policy and Government, George Mason University, Arlington, Virginia, USA
- VA National Center on Homelessness Among Veterans, Durham, North Carolina, USA
| | - Mark A Ilgen
- Department of Psychiatry, University of Michigan, Ann Arbor, Michigan, USA
- VA Health Services Research & Development, Center for Clinical Management Research, VA Ann Arbor Health Care System, Ann Arbor, Michigan, USA
| | - Ruth M Reeves
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
- VA Health Sevices Research & Development, VA Tennessee Valley Health Care System, Nashville, Tennessee, USA
| | - Alex H S Harris
- VA Health Services Research & Development, Center for Innovation to Implementation, VA Palo Alto Health Care System, Palo Alto, California, USA
- Stanford-Surgical Policy Improvement Research and Education Center, Stanford University School of Medicine, Stanford, California, USA
| |
Collapse
|
10
|
Natural Language Processing in Radiology: Update on Clinical Applications. J Am Coll Radiol 2022; 19:1271-1285. [PMID: 36029890 DOI: 10.1016/j.jacr.2022.06.016] [Citation(s) in RCA: 18] [Impact Index Per Article: 9.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/21/2022] [Revised: 05/25/2022] [Accepted: 06/03/2022] [Indexed: 11/24/2022]
Abstract
Radiological reports are a valuable source of information used to guide clinical care and support research. Organizing and managing this content, however, frequently requires several manual curations due to the more common unstructured nature of the reports. However, manual review of these reports for clinical knowledge extraction is costly and time-consuming. Natural language processing (NLP) is a set of methods developed to extract structured meaning from a body of text and can be used to optimize the workflow of health care professionals. Specifically, NLP methods can help radiologists as decision support systems and improve the management of patients' medical data. In this study, we highlight the opportunities offered by NLP in the field of radiology. A comprehensive review of the most commonly used NLP methods to extract information from radiological reports and the development of tools to improve radiological workflow using this information is presented. Finally, we review the important limitations of these tools and discuss the relevant observations and trends in the application of NLP to radiology that could benefit the field in the future.
Collapse
|
11
|
Xavier BA, Chen PH. Natural Language Processing for Imaging Protocol Assignment: Machine Learning for Multiclass Classification of Abdominal CT Protocols Using Indication Text Data. J Digit Imaging 2022; 35:1120-1130. [PMID: 35654878 PMCID: PMC9582109 DOI: 10.1007/s10278-022-00633-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/24/2021] [Revised: 03/09/2022] [Accepted: 04/05/2022] [Indexed: 11/25/2022] Open
Abstract
A correct protocol assignment is critical to high-quality imaging examinations, and its automation can be amenable to natural language processing (NLP). Assigning protocols for abdominal imaging CT scans is particularly challenging given the multiple organ specific indications and parameters. We compared conventional machine learning, deep learning, and automated machine learning builder workflows for this multiclass text classification task. A total of 94,501 CT studies performed over 4 years and their assigned protocols were obtained. Text data associated with each study including the ordering provider generated free text study indication and ICD codes were used for NLP analysis and protocol class prediction. The data was classified into one of 11 abdominal CT protocol classes before and after augmentations used to account for imbalances in the class sample sizes. Four machine learning (ML) algorithms, one deep learning algorithm, and an automated machine learning (AutoML) builder were used for the multilabel classification task: Random Forest (RF), Tree Ensemble (TE), Gradient Boosted Tree (GBT), multi-layer perceptron (MLP), Universal Language Model Fine-tuning (ULMFiT), and Google’s AutoML builder (Alphabet, Inc., Mountain View, CA), respectively. On the unbalanced dataset, the manually coded algorithms all performed similarly with F1 scores of 0.811 for RF, 0.813 for TE, 0.813 for GBT, 0.828 for MLP, and 0.847 for ULMFiT. The AutoML builder performed better with a F1 score of 0.854. On the balanced dataset, the tree ensemble machine learning algorithm performed the best with an F1 score of 0.803 and a Cohen’s kappa of 0.612. AutoML methods took a longer time for completion of NLP model training and evaluation, 4 h and 45 min compared to an average of 51 min for manual methods. Machine learning and natural language processing can be used for the complex multiclass classification task of abdominal imaging CT scan protocol assignment.
Collapse
Affiliation(s)
- Brian Arun Xavier
- Imaging Institute, Cleveland Clinic Foundation, 9500 Euclid Ave., P34, Cleveland, OH, 44195, USA.
| | - Po-Hao Chen
- Imaging Institute, Cleveland Clinic Foundation, 9500 Euclid Ave., P34, Cleveland, OH, 44195, USA
| |
Collapse
|
12
|
Linna N, Kahn CE. Applications of Natural Language Processing in Radiology: A Systematic Review. Int J Med Inform 2022; 163:104779. [DOI: 10.1016/j.ijmedinf.2022.104779] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/07/2022] [Revised: 03/28/2022] [Accepted: 04/21/2022] [Indexed: 12/27/2022]
|
13
|
Donnelly LF, Grzeszczuk R, Guimaraes CV. Use of Natural Language Processing (NLP) in Evaluation of Radiology Reports: An Update on Applications and Technology Advances. Semin Ultrasound CT MR 2022; 43:176-181. [PMID: 35339258 DOI: 10.1053/j.sult.2022.02.007] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/11/2022]
Abstract
Natural language processing (NLP) is focused on the computer interpretation of human language and can be used to evaluate radiology reports and has demonstrated useful applications in essentially all aspects of medical imaging delivery: interpretation of imaging data, improving image acquisition, image analysis, and increasing efficiency of imaging services. This manuscript reviews general technologic approaches to NLP at a level hopefully understandable by clinical radiologists, discusses recent advancements in NLP techniques, and discusses current and potential applications of NLP in radiology.
Collapse
Affiliation(s)
- Lane F Donnelly
- University of North Carolina, School of Medicine, Department of Radiology, Chapel Hill, NC; Stanford University, School of Medicine, Department of Radiology, Palo Alto, CA; Stanford University, School of Medicine, Department of Pediatrics, Palo Alto, CA.
| | | | - Carolina V Guimaraes
- University of North Carolina, School of Medicine, Department of Radiology, Chapel Hill, NC; Stanford University, School of Medicine, Department of Radiology, Palo Alto, CA
| |
Collapse
|
14
|
Crombé A, Seux M, Bratan F, Bergerot JF, Banaste N, Thomson V, Lecomte JC, Gorincour G. What Influences the Way Radiologists Express Themselves in Their Reports? A Quantitative Assessment Using Natural Language Processing. J Digit Imaging 2022; 35:993-1007. [PMID: 35318544 PMCID: PMC8939885 DOI: 10.1007/s10278-022-00619-6] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/14/2021] [Revised: 03/07/2022] [Accepted: 03/09/2022] [Indexed: 11/29/2022] Open
Abstract
Although using standardized reports is encouraged, most emergency radiological reports in France remain in free-text format that can be mined with natural language processing for epidemiological purposes, activity monitoring or data collection. These reports are obtained under various on-call conditions by radiologists with various backgrounds. Our aim was to investigate what influences the radiologists’ written expressions. To do so, this retrospective multicentric study included 30,227 emergency radiological reports of computed tomography scans and magnetic resonance imaging involving exactly one body region, only with pathological findings, interpreted from 2019–09-01 to 2020–02-28 by 165 radiologists. After text pre-processing, one-word tokenization and use of dictionaries for stop words, polarity, sentiment and uncertainty, 11 variables depicting the structure and content of words and sentences in the reports were extracted and summarized to 3 principal components capturing 93.7% of the dataset variance. In multivariate analysis, the 1st principal component summarized the length and lexical diversity of the reports and was significantly influenced by the weekday, time slot, workload, number of examinations previously interpreted by the radiologist during the on-call period, type of examination, emergency level and radiologists’ gender (P value range: < 0.0001–0.0029). The 2nd principal component summarized negative formulations, polarity and sentence length and was correlated with the number of examination previously interpreted by the radiologist, type of examination, emergency level, imaging modality and radiologists’ experience (P value range: < 0.0001–0.0032). The last principal component summarized questioning, uncertainty and polarity and was correlated with the type of examination and emergency level (all P values < 0.0001). Thus, the length, structure and content of emergency radiological reports were significantly influenced by organizational, radiologist- and examination-related characteristics, highlighting the subjectivity and variability in the way radiologists express themselves during their clinical activity. These findings advocate for more homogeneous practices in radiological reporting and stress the need to consider these influential features when developing models based on natural language processing.
Collapse
Affiliation(s)
- Amandine Crombé
- IMADIS, 48 rue quivogne, 63002, Lyon, France. .,University of Bordeaux, 33000, Bordeaux, France.
| | - Mylène Seux
- IMADIS, 48 rue quivogne, 63002, Lyon, France
| | - Flavie Bratan
- IMADIS, 48 rue quivogne, 63002, Lyon, France.,Department of Diagnostic and Interventional Imaging, Centre Hospitalier Saint-Joseph Saint-Luc, 69007, Lyon, France
| | - Jean-François Bergerot
- IMADIS, 48 rue quivogne, 63002, Lyon, France.,Ramsay Générale de Santé, Clinique Convert, 01000, Bourg-en-Bresse, France
| | - Nathan Banaste
- IMADIS, 48 rue quivogne, 63002, Lyon, France.,Department of Radiology, Hôpital Nord-Ouest, 69400, Villefranche-sur-Saône, France
| | - Vivien Thomson
- IMADIS, 48 rue quivogne, 63002, Lyon, France.,Ramsay Générale de Santé, Clinique de la Sauvegarde, 69009, Lyon, France
| | - Jean-Christophe Lecomte
- IMADIS, 48 rue quivogne, 63002, Lyon, France.,Centre Hospitalier de Saintonge, 17100, Saintes, France.,Centre Aquitain d'Imagerie, 33600, Pessac, France
| | - Guillaume Gorincour
- IMADIS, 48 rue quivogne, 63002, Lyon, France.,ELSAN, Clinique Bouchard, 13006, Marseille, France
| |
Collapse
|
15
|
Middelberg LK, Leonard JC, Shi J, Aranda A, Brown JC, Cochran CL, Eastep K, Gonzalez R, Haasz M, Herskovitz S, Hoffmann JA, Koral A, Lamoshi A, Levitte S, Lo YHJ, Montminy T, Novak I, Ng K, Novotny NM, Parrado RH, Ruan W, Shapiro J, Sinclair EM, Stewart AM, Talathi S, Tavarez MM, Townsend P, Zaytsev J, Rudolph B. High-Powered Magnet Exposures in Children: A Multi-Center Cohort Study. Pediatrics 2022; 149:184737. [PMID: 35112127 DOI: 10.1542/peds.2021-054543] [Citation(s) in RCA: 12] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Accepted: 11/23/2021] [Indexed: 01/08/2023] Open
Abstract
BACKGROUND AND OBJECTIVES High-powered magnets were effectively removed from the US market by the Consumer Product Safety Commission (CPSC) in 2012 but returned in 2016 after federal court decisions. The United States Court of Appeals for the 10th Circuit cited imprecise data among other reasons as justification for overturning CPSC protections. Since then, incidence of high-powered magnet exposure has increased markedly, but outcome data are limited. In this study, we aim to describe the epidemiology and outcomes in children seeking medical care for high-powered magnets after reintroduction to market. METHODS This is a multicenter, retrospective cohort study of patients aged 0 to 21 years with a confirmed high-powered magnet exposure (ie, ingestion or insertion) at 25 children's hospitals in the United States between 2017 and 2019. RESULTS Of 596 patients with high-powered magnet exposures identified, 362 (60.7%) were male and 566 (95%) were <14 years of age. Nearly all sought care for magnet ingestion (n = 574, 96.3%), whereas 17 patients (2.9%) presented for management of nasal or aural magnet foreign bodies, 4 (0.7%) for magnets in their genitourinary tract, and 1 patient (0.2%) had magnets in their respiratory tract. A total of 57 children (9.6%) had a life-threatening morbidity; 276 (46.3%) required an endoscopy, surgery, or both; and 332 (55.7%) required hospitalization. There was no reported mortality. CONCLUSIONS Despite being intended for use by those >14 years of age, high-powered magnets frequently cause morbidity and lead to high need for invasive intervention and hospitalization in children of all ages.
Collapse
Affiliation(s)
- Leah K Middelberg
- Department of Pediatrics, Division of Emergency Medicine, Nationwide Children's Hospital, The Ohio State University College of Medicine, Columbus, Ohio
| | - Julie C Leonard
- Department of Pediatrics, Division of Emergency Medicine, Nationwide Children's Hospital, The Ohio State University College of Medicine, Columbus, Ohio
| | - Junxin Shi
- Department of Pediatrics, Division of Emergency Medicine, Nationwide Children's Hospital, The Ohio State University College of Medicine, Columbus, Ohio
| | - Arturo Aranda
- Division of Pediatric Surgery, Dayton Children's Hospital, Dayton, Ohio
| | - Julie C Brown
- Seattle Children's Hospital, Department of Pediatrics, Division of Emergency Medicine, Seattle, Washington
| | - Christina L Cochran
- Department of Pediatrics, Division of Emergency Medicine, Children's of Alabama, University of Alabama at Birmingham College of Medicine, Birmingham, Alabama
| | - Kasi Eastep
- Department of Pediatrics, Division of Pediatric Emergency Medicine, Norton Children's Hospital affiliated with University of Louisville School of Medicine, Louisville, Kentucky
| | - Raquel Gonzalez
- Division of Pediatric Surgery, Johns Hopkins All Children's Hospital, Saint Petersburg, Florida
| | - Maya Haasz
- Department of APediatrics, Section of Pediatric Emergency Medicine, Children's Hospital Colorado, University of Colorado School of Medicine, Aurora, Colorado
| | - Scott Herskovitz
- Department of Pediatrics, Division of Emergency Medicine, Rady Children's Hospital, San Diego, California
| | - Jennifer A Hoffmann
- Division of Emergency Medicine, Ann & Robert H. Lurie Children's Hospital of Chicago, Northwestern University Feinberg School of Medicine, Chicago, Illinois
| | - Alexander Koral
- Department of Pediatrics, Section of Pediatric Gastroenterology and Hepatology, Yale New Haven Children's Hospital, Yale School of Medicine, New Haven, Connecticut
| | - Abdulraouf Lamoshi
- Division of Pediatric Surgery, Cohen Children's Medical Center; Northwell Health, Queens, New York
| | - Steven Levitte
- Division of Pediatric Gastroenterology, Hepatology, and Nutrition, Lucile Packard Children's Hospital, Stanford University, Palo Alto, California
| | - Yu Hsiang J Lo
- Department of Emergency Medicine, C.S. Mott Children's Hospital, University of Michigan, Ann Arbor, Michigan
| | - Taylor Montminy
- Department of Pediatrics, Children's Hospital Colorado, University of Colorado School of Medicine, Aurora, Colorado
| | - Inna Novak
- Children's Hospital at Montefiore, Albert Einstein College of Medicine; Bronx, New York
| | - Kenneth Ng
- Department of Pediatrics, Division of Pediatric Gastroenterology, Hepatology, and Nutrition, Johns Hopkins Children's Center, Johns Hopkins University School of Medicine, Baltimore, Maryland
| | - Nathan M Novotny
- Beaumont Children's, Oakland University William Beaumont School of Medicine, Royal Oak, Michigan
| | - Raphael H Parrado
- Division of Pediatric Surgery, Department of Surgery, Medical University of South Carolina Shawn Jenkins Children's Hospital, Charleston, South Carolina
| | - Wenly Ruan
- Department of Pediatrics, Section of Pediatric Gastroenterology, Hepatology, and Nutrition, Texas Children's Hospital, Baylor College of Medicine, Houston, Texas
| | - Joseph Shapiro
- Division of Emergency Medicine, Children's National Hospital, Washington, District of Columbia
| | - Elizabeth M Sinclair
- Pediatric Gastroenterology, Hepatology, and Nutrition, Children's Healthcare of Atlanta, Emory University School of Medicine, Department of Pediatrics, Atlanta, Georgia
| | - Amanda M Stewart
- Department of Pediatrics, Division of Emergency Medicine, Boston Children's Hospital, Boston, Massachusetts
| | - Saurabh Talathi
- Department of Pediatrics, Division of Pediatric Gastroenterology, Hepatology, and Nutrition, Oklahoma Children's Hospital, University of Oklahoma College of Medicine, Oklahoma City, Oklahoma
| | - Melissa M Tavarez
- Department of Pediatrics, UPMC Children's Hospital of Pittsburgh, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania
| | - Peter Townsend
- Department of Pediatrics, Division of Gastroenterology, Connecticut Children's Hospital, University of Connecticut School of Medicine, Hartford, Connecticut
| | - Julia Zaytsev
- University of Texas Southwestern Medical School, Dallas, Texas
| | - Bryan Rudolph
- Children's Hospital at Montefiore, Albert Einstein College of Medicine; Bronx, New York
| |
Collapse
|
16
|
Filice RW, Kahn CE. Biomedical Ontologies to Guide AI Development in Radiology. J Digit Imaging 2021; 34:1331-1341. [PMID: 34724143 PMCID: PMC8669056 DOI: 10.1007/s10278-021-00527-1] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/29/2021] [Revised: 04/27/2021] [Accepted: 10/13/2021] [Indexed: 10/25/2022] Open
Abstract
The advent of deep learning has engendered renewed and rapidly growing interest in artificial intelligence (AI) in radiology to analyze images, manipulate textual reports, and plan interventions. Applications of deep learning and other AI approaches must be guided by sound medical knowledge to assure that they are developed successfully and that they address important problems in biomedical research or patient care. To date, AI has been applied to a limited number of real-world radiology applications. As AI systems become more pervasive and are applied more broadly, they will benefit from medical knowledge on a larger scale, such as that available through computer-based approaches. A key approach to represent computer-based knowledge in a particular domain is an ontology. As defined in informatics, an ontology defines a domain's terms through their relationships with other terms in the ontology. Those relationships, then, define the terms' semantics, or "meaning." Biomedical ontologies commonly define the relationships between terms and more general terms, and can express causal, part-whole, and anatomic relationships. Ontologies express knowledge in a form that is both human-readable and machine-computable. Some ontologies, such as RSNA's RadLex radiology lexicon, have been applied to applications in clinical practice and research, and may be familiar to many radiologists. This article describes how ontologies can support research and guide emerging applications of AI in radiology, including natural language processing, image-based machine learning, radiomics, and planning.
Collapse
Affiliation(s)
- Ross W Filice
- Department of Radiology, MedStar Georgetown University Hospital, Washington, DC, USA
| | - Charles E Kahn
- Department of Radiology and Institute for Biomedical Informatics, University of Pennsylvania, 3400 Spruce Street, Philadelphia, PA, 19104, USA.
| |
Collapse
|
17
|
Bizzo BC, Almeida RR, Alkasab TK. Artificial Intelligence Enabling Radiology Reporting. Radiol Clin North Am 2021; 59:1045-1052. [PMID: 34689872 DOI: 10.1016/j.rcl.2021.07.004] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/26/2022]
Abstract
The radiology reporting process is beginning to incorporate structured, semantically labeled data. Tools based on artificial intelligence technologies using a structured reporting context can assist with internal report consistency and longitudinal tracking. To-do lists of relevant issues could be assembled by artificial intelligence tools, incorporating components of the patient's history. Radiologists will review and select artificial intelligence-generated and other data to be transmitted to the electronic health record and generate feedback for ongoing improvement of artificial intelligence tools. These technologies should make reports more valuable by making reports more accessible and better able to integrate into care pathways.
Collapse
Affiliation(s)
- Bernardo C Bizzo
- Department of Radiology, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, Founders 210, Boston, MA 02114, USA
| | - Renata R Almeida
- Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, 75 Francis St, Boston, MA 02115, USA
| | - Tarik K Alkasab
- Department of Radiology, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, Founders 210, Boston, MA 02114, USA.
| |
Collapse
|
18
|
Steinkamp J, Cook TS. Basic Artificial Intelligence Techniques: Natural Language Processing of Radiology Reports. Radiol Clin North Am 2021; 59:919-931. [PMID: 34689877 DOI: 10.1016/j.rcl.2021.06.003] [Citation(s) in RCA: 6] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/20/2022]
Abstract
Natural language processing (NLP) is a subfield of computer science and linguistics that can be applied to extract meaningful information from radiology reports. Symbolic NLP is rule based and well suited to problems that can be explicitly defined by a set of rules. Statistical NLP is better situated to problems that cannot be well defined and requires annotated or labeled examples from which machine learning algorithms can infer the rules. Both symbolic and statistical NLP have found success in a variety of radiology use cases. More recently, deep learning approaches, including transformers, have gained traction and demonstrated good performance.
Collapse
Affiliation(s)
- Jackson Steinkamp
- Department of Medicine, Hospital of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, PA 19104, USA
| | - Tessa S Cook
- Perelman School of Medicine at the University of Pennsylvania, 3400 Spruce Street, 1 Silverstein Radiology, Philadelphia, PA 19104, USA.
| |
Collapse
|
19
|
Olthof AW, van Ooijen PMA, Cornelissen LJ. Deep Learning-Based Natural Language Processing in Radiology: The Impact of Report Complexity, Disease Prevalence, Dataset Size, and Algorithm Type on Model Performance. J Med Syst 2021; 45:91. [PMID: 34480231 PMCID: PMC8416876 DOI: 10.1007/s10916-021-01761-4] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/30/2021] [Accepted: 08/04/2021] [Indexed: 12/12/2022]
Abstract
In radiology, natural language processing (NLP) allows the extraction of valuable information from radiology reports. It can be used for various downstream tasks such as quality improvement, epidemiological research, and monitoring guideline adherence. Class imbalance, variation in dataset size, variation in report complexity, and algorithm type all influence NLP performance but have not yet been systematically and interrelatedly evaluated. In this study, we investigate these factors on the performance of four types [a fully connected neural network (Dense), a long short-term memory recurrent neural network (LSTM), a convolutional neural network (CNN), and a Bidirectional Encoder Representations from Transformers (BERT)] of deep learning-based NLP. Two datasets consisting of radiologist-annotated reports of both trauma radiographs (n = 2469) and chest radiographs and computer tomography (CT) studies (n = 2255) were split into training sets (80%) and testing sets (20%). The training data was used as a source to train all four model types in 84 experiments (Fracture-data) and 45 experiments (Chest-data) with variation in size and prevalence. The performance was evaluated on sensitivity, specificity, positive predictive value, negative predictive value, area under the curve, and F score. After the NLP of radiology reports, all four model-architectures demonstrated high performance with metrics up to > 0.90. CNN, LSTM, and Dense were outperformed by the BERT algorithm because of its stable results despite variation in training size and prevalence. Awareness of variation in prevalence is warranted because it impacts sensitivity and specificity in opposite directions.
Collapse
Affiliation(s)
- A W Olthof
- Department of Radiation Oncology, University of Groningen, University Medical Center Groningen, Hanzeplein 1, Groningen, The Netherlands. .,Treant Health Care Group, Department of Radiology, Dr G.H. Amshoffweg 1, Hoogeveen, The Netherlands. .,Hospital Group Twente (ZGT), Department of Radiology, Almelo, The Netherlands.
| | - P M A van Ooijen
- Department of Radiation Oncology, University of Groningen, University Medical Center Groningen, Hanzeplein 1, Groningen, The Netherlands.,Data Science Center in Health (DASH), University of Groningen, University Medical Center Groningen, Machine Learning Lab, L.J, Zielstraweg 2, Groningen, The Netherlands
| | - L J Cornelissen
- Department of Radiation Oncology, University of Groningen, University Medical Center Groningen, Hanzeplein 1, Groningen, The Netherlands.,COSMONiO Imaging BV, L.J, Zielstraweg 2, Groningen, The Netherlands
| |
Collapse
|
20
|
Richardson ML, Garwood ER, Lee Y, Li MD, Lo HS, Nagaraju A, Nguyen XV, Probyn L, Rajiah P, Sin J, Wasnik AP, Xu K. Noninterpretive Uses of Artificial Intelligence in Radiology. Acad Radiol 2021; 28:1225-1235. [PMID: 32059956 DOI: 10.1016/j.acra.2020.01.012] [Citation(s) in RCA: 49] [Impact Index Per Article: 16.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/17/2019] [Revised: 01/08/2020] [Accepted: 01/09/2020] [Indexed: 12/12/2022]
Abstract
We deem a computer to exhibit artificial intelligence (AI) when it performs a task that would normally require intelligent action by a human. Much of the recent excitement about AI in the medical literature has revolved around the ability of AI models to recognize anatomy and detect pathology on medical images, sometimes at the level of expert physicians. However, AI can also be used to solve a wide range of noninterpretive problems that are relevant to radiologists and their patients. This review summarizes some of the newer noninterpretive uses of AI in radiology.
Collapse
Affiliation(s)
| | - Elisabeth R Garwood
- Department of Radiology, University of Massachusetts, Worcester, Massachusetts
| | - Yueh Lee
- Department of Radiology, University of North Carolina, Chapel Hill, North Carolina
| | - Matthew D Li
- Department of Radiology, Harvard Medical School/Massachusetts General Hospital, Boston, Massachusets
| | - Hao S Lo
- Department of Radiology, University of Washington, Seattle, Washington
| | - Arun Nagaraju
- Department of Radiology, University of Chicago, Chicago, Illinois
| | - Xuan V Nguyen
- Department of Radiology, The Ohio State University Wexner Medical Center, Columbus, Ohio
| | - Linda Probyn
- Department of Radiology, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, Ontario
| | - Prabhakar Rajiah
- Department of Radiology, University of Texas Southwestern Medical Center, Dallas, Texas
| | - Jessica Sin
- Department of Radiology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
| | - Ashish P Wasnik
- Department of Radiology, University of Michigan, Ann Arbor, Michigan
| | - Kali Xu
- Department of Medicine, Santa Clara Valley Medical Center, Santa Clara, California
| |
Collapse
|
21
|
Olthof AW, Shouche P, Fennema EM, IJpma FFA, Koolstra RHC, Stirler VMA, van Ooijen PMA, Cornelissen LJ. Machine learning based natural language processing of radiology reports in orthopaedic trauma. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2021; 208:106304. [PMID: 34333208 DOI: 10.1016/j.cmpb.2021.106304] [Citation(s) in RCA: 15] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 05/22/2020] [Accepted: 07/18/2021] [Indexed: 06/13/2023]
Abstract
OBJECTIVES To compare different Machine Learning (ML) Natural Language Processing (NLP) methods to classify radiology reports in orthopaedic trauma for the presence of injuries. Assessing NLP performance is a prerequisite for downstream tasks and therefore of importance from a clinical perspective (avoiding missed injuries, quality check, insight in diagnostic yield) as well as from a research perspective (identification of patient cohorts, annotation of radiographs). METHODS Datasets of Dutch radiology reports of injured extremities (n = 2469, 33% fractures) and chest radiographs (n = 799, 20% pneumothorax) were collected in two different hospitals and labeled by radiologists and trauma surgeons for the presence or absence of injuries. NLP classification was applied and optimized by testing different preprocessing steps and different classifiers (Rule-based, ML, and Bidirectional Encoder Representations from Transformers (BERT)). Performance was assessed by F1-score, AUC, sensitivity, specificity and accuracy. RESULTS The deep learning based BERT model outperforms all other classification methods which were assessed. The model achieved an F1-score of (95 ± 2)% and accuracy of (96 ± 1)% on a dataset of simple reports (n= 2469), and an F1 of (83 ± 7)% with accuracy (93 ± 2)% on a dataset of complex reports (n= 799). CONCLUSION BERT NLP outperforms traditional ML and rule-base classifiers when applied to Dutch radiology reports in orthopaedic trauma.
Collapse
Affiliation(s)
- A W Olthof
- Department of Radiology, Treant Health Care Group, Dr. G.H. Amshoffweg 1, Hoogeveen, the Netherlands; Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands.
| | - P Shouche
- Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands
| | - E M Fennema
- Department of Trauma Surgery, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands
| | - F F A IJpma
- Department of Trauma Surgery, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands
| | - R H C Koolstra
- Department of Radiology, Treant Health Care Group, Dr. G.H. Amshoffweg 1, Hoogeveen, the Netherlands
| | - V M A Stirler
- Department of Trauma Surgery, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands
| | - P M A van Ooijen
- Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands; Machine Learning Lab, Data Science Center in Health (DASH),University Medical Center Groningen, University of Groningen, L.J. Zielstraweg 2, Groningen, the Netherlands
| | - L J Cornelissen
- Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands; COSMONiO Imaging BV, L.J. Zielstraweg 2, Groningen, the Netherlands
| |
Collapse
|
22
|
Wiggins WF, Kitamura F, Santos I, Prevedello LM. Natural Language Processing of Radiology Text Reports: Interactive Text Classification. Radiol Artif Intell 2021; 3:e210035. [PMID: 34350414 DOI: 10.1148/ryai.2021210035] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/26/2021] [Revised: 04/15/2021] [Accepted: 04/22/2021] [Indexed: 11/11/2022]
Abstract
This report presents a hands-on introduction to natural language processing (NLP) of radiology reports with deep neural networks in Google Colaboratory (Colab) to introduce readers to the rapidly evolving field of NLP. The implementation of the Google Colab notebook was designed with code hidden to facilitate learning for noncoders (ie, individuals with little or no computer programming experience). The data used for this module are the corpus of radiology reports from the Indiana University chest x-ray collection available from the National Library of Medicine's Open-I service. The module guides learners through the process of exploring the data, splitting the data for model training and testing, preparing the data for NLP analysis, and training a deep NLP model to classify the reports as normal or abnormal. Concepts in NLP, such as tokenization, numericalization, language modeling, and word embeddings, are demonstrated in the module. The module is implemented in a guided fashion with the authors presenting the material and explaining concepts. Interactive features and extensive text commentary are provided directly in the notebook to facilitate self-guided learning and experimentation with the module. Keywords: Neural Networks, Negative Expression Recognition, Natural Language Processing, Computer Applications, Informatics © RSNA, 2021.
Collapse
Affiliation(s)
- Walter F Wiggins
- Department of Radiology, Duke University Health System, Duke University Hospital, Box 3808, 2301 Erwin Rd, Durham, NC 27710 (W.F.W.); Department of Diagnostic Imaging, Universidade Federal de São Paulo, Escola Paulista de Medicina, São Paulo, Brazil (F.K., I.S.); Head of AI, Diagnósticos da América SA (DASA), São Paulo, Brazil (F.K.); FIDI, NESS Health, São Paulo, Brazil (I.S.); and Department of Radiology, Ohio State University, Columbus, Ohio (L.M.P.)
| | - Felipe Kitamura
- Department of Radiology, Duke University Health System, Duke University Hospital, Box 3808, 2301 Erwin Rd, Durham, NC 27710 (W.F.W.); Department of Diagnostic Imaging, Universidade Federal de São Paulo, Escola Paulista de Medicina, São Paulo, Brazil (F.K., I.S.); Head of AI, Diagnósticos da América SA (DASA), São Paulo, Brazil (F.K.); FIDI, NESS Health, São Paulo, Brazil (I.S.); and Department of Radiology, Ohio State University, Columbus, Ohio (L.M.P.)
| | - Igor Santos
- Department of Radiology, Duke University Health System, Duke University Hospital, Box 3808, 2301 Erwin Rd, Durham, NC 27710 (W.F.W.); Department of Diagnostic Imaging, Universidade Federal de São Paulo, Escola Paulista de Medicina, São Paulo, Brazil (F.K., I.S.); Head of AI, Diagnósticos da América SA (DASA), São Paulo, Brazil (F.K.); FIDI, NESS Health, São Paulo, Brazil (I.S.); and Department of Radiology, Ohio State University, Columbus, Ohio (L.M.P.)
| | - Luciano M Prevedello
- Department of Radiology, Duke University Health System, Duke University Hospital, Box 3808, 2301 Erwin Rd, Durham, NC 27710 (W.F.W.); Department of Diagnostic Imaging, Universidade Federal de São Paulo, Escola Paulista de Medicina, São Paulo, Brazil (F.K., I.S.); Head of AI, Diagnósticos da América SA (DASA), São Paulo, Brazil (F.K.); FIDI, NESS Health, São Paulo, Brazil (I.S.); and Department of Radiology, Ohio State University, Columbus, Ohio (L.M.P.)
| |
Collapse
|
23
|
Masua B, Masasi N. Enhancing text pre-processing for Swahili language: Datasets for common Swahili stop-words, slangs and typos with equivalent proper words. Data Brief 2020; 33:106517. [PMID: 33294515 PMCID: PMC7689026 DOI: 10.1016/j.dib.2020.106517] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/14/2020] [Revised: 11/06/2020] [Accepted: 11/06/2020] [Indexed: 11/26/2022] Open
Abstract
Natural Language Processing requires data to be pre-processed to guarantee quality models in different machine learning tasks. However, Swahili language have been disadvantaged and is classified as low resource language because of inadequate data for NLP especially basic textual datasets that are useful during pre-processing stage. In this article we develop and contribute common Swahili Stop-words, common Swahili Slangs and common Swahili Typos datasets. The main source for these datasets were short Swahili messages collected from Tanzanian platform that is used by young people to convey their opinions on things that matters to them. Therefore, we derive list of common Swahili stop-words by reviewing most frequent words that are generated with Python script from our corpus, review common slang with help of Swahili experts with their corresponding proper words, and generate common Swahili typos by analysing least frequent words generated by a Python script from corpus. The datasets were exported into files for easy access and reuse. These datasets can be reused in natural language processing as resources in pre-processing phase for Swahili textual data.
Collapse
Affiliation(s)
- Bernard Masua
- College of Information and Communication Technologies (CoICT), University of Dar Es Salaam, Ali Hassan Mwinyi Road, Kijitonyama campus, Dar Es Salaam, TZ 33335, Tanzania
| | - Noel Masasi
- College of Information and Communication Technologies (CoICT), University of Dar Es Salaam, Ali Hassan Mwinyi Road, Kijitonyama campus, Dar Es Salaam, TZ 33335, Tanzania
| |
Collapse
|
24
|
Jha S, Cook T. Artificial Intelligence in Radiology--The State of the Future. Acad Radiol 2020; 27:1-2. [PMID: 31753720 DOI: 10.1016/j.acra.2019.11.003] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/10/2019] [Accepted: 11/10/2019] [Indexed: 12/18/2022]
Affiliation(s)
- Saurabh Jha
- Department of Radiology, MRI Learning Center Hospital, University of Pennsylvania, 3400 Spruce Street, Phila, PA 19104, United States.
| | - Tessa Cook
- Department of Radiology, MRI Learning Center Hospital, University of Pennsylvania, 3400 Spruce Street, Phila, PA 19104, United States
| |
Collapse
|