scholarly journals Labelling chest x-ray reports using an open-source NLP and ML tool for text data binary classification

2019 ◽  
Author(s):  
Sohrab Towfighi ◽  
Arnav Agarwal ◽  
Denise Y. F. Mak ◽  
Amol Verma
Keyword(s):  
Internal Medicine ◽  
Language Processing ◽  
Performance Metrics ◽  
Random Search ◽  
General Medicine ◽  
X Rays ◽  
Text Data ◽  
X Ray ◽  
Radiology Reports ◽  
Chest X Ray

AbstractThe chest x-ray is a commonly requested diagnostic test on internal medicine wards which can diagnose many acute pathologies needing intervention. We developed a natural language processing (NLP) and machine learning (ML) model to identify the presence of opacities or endotracheal intubation on chest x-rays using only the radiology report. This a preliminary report of our work and findings. Using the General Medicine Inpatient Initiative (GEMINI) dataset, housing inpatient clinical and administrative data from 7 major hospitals, we retrieved 1000 plain film radiology reports which were classified according to 4 labels by an internal medicine resident. NLP/ML models were developed to identify the following on the radiograph reports: the report is that of a chest x-ray, there is definite absence of an opacity, there is definite presence of an opacity, the report is a follow-up report with minimal details in its text, and there is an endotracheal tube in place. Our NLP/ML model development methodology included a random search of either TF-IDF or bag-of-words for vectorization along with random search of various ML models. Our Python programming scripts were made publicly available on GitHub to allow other parties to train models using their own text data. 100 randomly generated ML pipelines were compared using 10-fold cross validation on 75% of the data, while 25% of the data was left out for generalizability testing. With respect to the question of whether a chest x-ray definitely lacks an opacity, the model’s performance metrics were accuracy of 0.84, precision of 0.94, recall of 0.81, and receiver operating characteristic area under curve of 0.86. Model performance was worse when trained against a highly imbalanced dataset despite the use of an advanced oversampling technique.

scholarly journals Objective Evaluation of Deep Uncertainty Predictions for COVID-19 Detection

2021 ◽  
Author(s):  
Hamzeh Asgharnezhad ◽  
Afshar Shamsi ◽  
Roohallah Alizadehsani ◽  
Abbas Khosravi ◽  
Saeid Nahavandi ◽  
...  
Keyword(s):  
Uncertainty Quantification ◽  
Performance Metrics ◽  
Confusion Matrix ◽  
Objective Evaluation ◽  
X Rays ◽  
The Novel ◽  
X Ray ◽  
Uncertainty Estimates ◽  
Chest X Ray ◽  
Novel Concept

Abstract Deep neural networks (DNNs) have been widely applied for detecting COVID-19 in medical images. Existing studies mainly apply transfer learning and other data representation strategies to generate accurate point estimates. The generalization power of these networks is always questionable due to being developed using small datasets and failing to report their predictive confidence. Quantifying uncertainties associated with DNN predictions is a prerequisite for their trusted deployment in medical settings. Here we apply and evaluate three uncertainty quantification techniques for COVID-19 detection using chest X-Ray (CXR) images. The novel concept of uncertainty confusion matrix is proposed and new performance metrics for the objective evaluation of uncertainty estimates are introduced. Through comprehensive experiments, it is shown that networks pertained on CXR images outperform networks pretrained on natural image datasets such as ImageNet. Qualitatively and quantitatively evaluations also reveal that the predictive uncertainty estimates are statistically higher for erroneous predictions than correct predictions. Accordingly, uncertainty quantification methods are capable of flagging risky predictions with high uncertainty estimates. We also observe that ensemble methods more reliably capture uncertainties during the inference. DNN-based solutions for COVID-19 detection have been mainly proposed without any principled mechanism for risk mitigation. Previous studies have mainly focused on on generating single-valued predictions using pretrained DNNs. In this paper, we comprehensively apply and comparatively evaluate three uncertainty quantification techniques for COVID-19 detection using chest X-Ray images. The novel concept of uncertainty confusion matrix is proposed and new performance metrics for the objective evaluation of uncertainty estimates are introduced for the first time. Using these new uncertainty performance metrics, we quantitatively demonstrate where and when we could trust DNN predictions for COVID-19 detection from chest X-rays. It is important to note the proposed novel uncertainty evaluation metrics are generic and could be applied for evaluation of probabilistic forecasts in all classification problems.

scholarly journals The Effect of Image Resolution on Automated Classification of Chest X-rays

2021 ◽  
Author(s):  
Md Inzamam Ul Haque ◽  
Abhishek K Dubey ◽  
Jacob D Hinkle
Keyword(s):  
High Resolution ◽  
Automated Analysis ◽  
Image Resolution ◽  
Free Text ◽  
X Rays ◽  
Training Time ◽  
X Ray ◽  
Radiology Reports ◽  
Chest X Ray ◽  
High Resolution Images

Deep learning models have received much attention lately for their ability to achieve expert-level performance on the accurate automated analysis of chest X-rays. Although publicly available chest X-ray datasets include high resolution images, most models are trained on reduced size images due to limitations on GPU memory and training time. As compute capability continues to advance, it will become feasible to train large convolutional neural networks on high-resolution images. This study is based on the publicly available MIMIC-CXR-JPG dataset, comprising 377,110 high resolution chest X-ray images, and provided with 14 labels to the corresponding free-text radiology reports. We find, interestingly, that tasks that require a large receptive field are better suited to downscaled input images, and we verify this qualitatively by inspecting effective receptive fields and class activation maps of trained models. Finally, we show that stacking an ensemble across resolutions outperforms each individual learner at all input resolutions while providing interpretable scale weights, suggesting that multi-scale features are crucially important to information extraction from high-resolution chest X-rays.

scholarly journals 1184. Making Pneumonia Surveillance Easy: Automation of Pneumonia Case Detection

2019 ◽  
Vol 6 (Supplement_2) ◽  
pp. S424-S425
Author(s):  
Dan Ding ◽  
Anna Stachel ◽  
Eduardo Iturrate ◽  
Michael Phillips
Keyword(s):  
Language Processing ◽  
Cost Effective ◽  
The United States ◽  
Surveillance Data ◽  
X Rays ◽  
Healthcare Facilities ◽  
X Ray ◽  
Chest X Ray ◽  
Sensitivity Specificity ◽  
Time Critical

Abstract Background Pneumonia (PNU) is the second most common nosocomial infection in the United States and is associated with substantial morbidity and mortality. While definitions from CDC were developed to increase the reliability of surveillance data, reduce the burden of surveillance in healthcare facilities, and enhance the utility of surveillance data for improving patient safety - the algorithm is still laborious. We propose an implementation of a refined algorithm script which combines two CDC definitions with the use of natural language processing (NLP), a tool which relies on pattern matching to determine whether a condition of interest is reported as present or absent in a report, to automate PNU surveillance. Methods Using SAS v9.4 to write a query, we used a combination of National Healthcare Safety Network’s (NHSN) PNU and ventilator-associated event (VAE) definitions that use discrete fields found in electronic medical records (EMR) and trained an NLP tool to determine whether chest x-ray report was indicative of PNU (Fig1). To validate, we assessed sensitivity/specificity of NLP tool results compared with clinicians’ interpretations. Results The NLP tool was highly accurate in classifying the presence of PNU in chest x-rays. After training the NLP tool, there were only 4% discrepancies between NLP tool and clinicians interpretations of 223 x-ray reports - sensitivity 92.2% (81.1–97.8), specificity 97.1% (93.4–99.1), PPV 90.4% (79.0–96.8), NPV 97.7% (94.1–99.4). Combining the automated use of discrete EMR fields with NLP tool significantly reduces the time spent manually reviewing EMRs. A manual review for PNU without automation requires approximately 10 minutes each day per admission. With a monthly average of 2,350 adult admissions at our hospital and 16,170 patient-days for admissions with at least 2 days, the algorithm saves approximately 2,695 review hours. Conclusion The use of discrete EMR fields with an NLP tool proves to be a timelier, cost-effective yet accurate alternative to manual PNU surveillance review. By allowing an automated algorithm to review PNU, timely reports can be sent to units about individual cases. Compared with traditional CDC surveillance definitions, an automated tool allows real-time critical review for infection and prevention activities. Disclosures All authors: No reported disclosures.

Development of a System that Generates Structured Reports for Chest X-ray Radiography

2010 ◽  
Vol 49 (04) ◽  
pp. 360-370 ◽  
Author(s):  
Y. Matsumura ◽  
N. Mihara ◽  
Y. Kawakami ◽  
K. Sasai ◽  
H. Takeda ◽  
...  
Keyword(s):  
Narrative Form ◽  
Free Text ◽  
X Rays ◽  
Sentence Structure ◽  
X Ray ◽  
Radiology Reports ◽  
Chest X Ray ◽  
Structured Reports ◽  
Description Framework ◽  
Resource Description

Summary Objectives: Radiology reports are typically made in narrative form; this is a barrier to the implementation of advanced applications for data analysis or a decision support. We developed a system that generates structured reports for chest x-ray radiography. Methods: Based on analyzing existing reports, we determined the fundamental sentence structure of findings as compositions of procedure, region, finding, and diagnosis. We categorized the observation objects into lung, mediastinum, bone, soft tissue, and pleura and chest wall. The terms of region, finding, and diagnosis were associated with each other. We expressed the terms and the relations between the terms using a resource description framework (RDF) and developed a reporting system based on it. The system shows a list of terms in each category, and modifiers can be entered using templates that are linked to each term. This system guides users to select terms by highlighting associated terms. Fifty chest x-rays with abnormal findings were interpreted by five radiologists and reports were made either by the system or by the free-text method. Results: The system decreased the time needed to make a report by 12.5% compared with the free-text method, and the sentences generated by the system were well concordant with those made by free-text method (F-measure = 90%). The results of the questionnaire showed that our system is applicable to radiology reports of chest x-rays in daily clinical practice. Conclusions: The method of generating structured reports for chest x-rays was feasible, because it generated almost concordant reports in shorter time compared with the free-text method.

Deep Neural Network-Based Screening Model for COVID-19-Infected Patients Using Chest X-Ray Images

2020 ◽  
pp. 2151004 ◽  
Author(s):  
Dilbag Singh ◽  
Vijay Kumar ◽  
Vaishali Yadav ◽  
Manjit Kaur
Keyword(s):  
Neural Network ◽  
Performance Metrics ◽  
Automated Analysis ◽  
X Rays ◽  
X Ray ◽  
Screening Model ◽  
X Ray Imaging ◽  
Adaptive Differential Evolution ◽  
Chest X Ray ◽  
Almost All

There are limited coronavirus disease 2019 (COVID-19) testing kits, therefore, development of other diagnosis approaches is desirable. The doctors generally utilize chest X-rays and Computed Tomography (CT) scans to diagnose pneumonia, lung inflammation, abscesses, and/or enlarged lymph nodes. Since COVID-19 attacks the epithelial cells that line our respiratory tract, therefore, X-ray images are utilized in this paper, to classify the patients with infected (COVID-19 [Formula: see text]ve) and uninfected (COVID-19 [Formula: see text]ve) lungs. Almost all hospitals have X-ray imaging machines, therefore, the chest X-ray images can be used to test for COVID-19 without utilizing any kind of dedicated test kits. However, the chest X-ray-based COVID-19 classification requires a radiology expert and significant time, which is precious when COVID-19 infection is increasing at a rapid rate. Therefore, the development of an automated analysis approach is desirable to save the medical professionals’ valuable time. In this paper, a deep convolutional neural network (CNN) approach is designed and implemented. Besides, the hyper-parameters of CNN are tuned using Multi-objective Adaptive Differential Evolution (MADE). Extensive experiments are performed by considering the benchmark COVID-19 dataset. Comparative analysis reveals that the proposed technique outperforms the competitive machine learning models in terms of various performance metrics.

scholarly journals Development and multicenter validation of chest X-ray radiography interpretations based on natural language processing

2021 ◽  
Vol 1 (1) ◽  
Author(s):  
Yaping Zhang ◽  
Mingqian Liu ◽  
Shundong Hu ◽  
Yao Shen ◽  
Jun Lan ◽  
...  
Keyword(s):  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Image Annotation ◽  
Training Dataset ◽  
Community Clinic ◽  
X Ray ◽  
Radiology Reports ◽  
Significant Difference ◽  
Chest X Ray

Abstract Background Artificial intelligence can assist in interpreting chest X-ray radiography (CXR) data, but large datasets require efficient image annotation. The purpose of this study is to extract CXR labels from diagnostic reports based on natural language processing, train convolutional neural networks (CNNs), and evaluate the classification performance of CNN using CXR data from multiple centers Methods We collected the CXR images and corresponding radiology reports of 74,082 subjects as the training dataset. The linguistic entities and relationships from unstructured radiology reports were extracted by the bidirectional encoder representations from transformers (BERT) model, and a knowledge graph was constructed to represent the association between image labels of abnormal signs and the report text of CXR. Then, a 25-label classification system were built to train and test the CNN models with weakly supervised labeling. Results In three external test cohorts of 5,996 symptomatic patients, 2,130 screening examinees, and 1,804 community clinic patients, the mean AUC of identifying 25 abnormal signs by CNN reaches 0.866 ± 0.110, 0.891 ± 0.147, and 0.796 ± 0.157, respectively. In symptomatic patients, CNN shows no significant difference with local radiologists in identifying 21 signs (p > 0.05), but is poorer for 4 signs (p < 0.05). In screening examinees, CNN shows no significant difference for 17 signs (p > 0.05), but is poorer at classifying nodules (p = 0.013). In community clinic patients, CNN shows no significant difference for 12 signs (p > 0.05), but performs better for 6 signs (p < 0.001). Conclusion We construct and validate an effective CXR interpretation system based on natural language processing.

scholarly journals Explainable Machine Learning for COVID-19 Pneumonia Classification With Texture-Based Features Extraction in Chest Radiography

2022 ◽  
Vol 3 ◽  
Author(s):  
Luís Vinícius de Moura ◽  
Christian Mattjie ◽  
Caroline Machado Dartora ◽  
Rodrigo C. Barros ◽  
Ana Maria Marques da Silva
Keyword(s):  
Machine Learning ◽  
Random Search ◽  
Texture Features ◽  
Classification Model ◽  
Recursive Feature Elimination ◽  
X Rays ◽  
X Ray ◽  
Reverse Transcription Pcr ◽  
Ensemble Machine Learning ◽  
Chest X Ray

Both reverse transcription-PCR (RT-PCR) and chest X-rays are used for the diagnosis of the coronavirus disease-2019 (COVID-19). However, COVID-19 pneumonia does not have a defined set of radiological findings. Our work aims to investigate radiomic features and classification models to differentiate chest X-ray images of COVID-19-based pneumonia and other types of lung patterns. The goal is to provide grounds for understanding the distinctive COVID-19 radiographic texture features using supervised ensemble machine learning methods based on trees through the interpretable Shapley Additive Explanations (SHAP) approach. We use 2,611 COVID-19 chest X-ray images and 2,611 non-COVID-19 chest X-rays. After segmenting the lung in three zones and laterally, a histogram normalization is applied, and radiomic features are extracted. SHAP recursive feature elimination with cross-validation is used to select features. Hyperparameter optimization of XGBoost and Random Forest ensemble tree models is applied using random search. The best classification model was XGBoost, with an accuracy of 0.82 and a sensitivity of 0.82. The explainable model showed the importance of the middle left and superior right lung zones in classifying COVID-19 pneumonia from other lung patterns.

scholarly journals COVIDXception-Net: A Bayesian Optimization Based Deep Learning Approach to Diagnose COVID-19 from X-Ray Images

2020 ◽  
Author(s):  
Shifat E Arman ◽  
Sejuti Rahman ◽  
Shamim Ahmed Deowan
Keyword(s):  
Loss Function ◽  
Random Search ◽  
Bayesian Optimization ◽  
X Rays ◽  
X Ray ◽  
Intelligence System ◽  
Proposed Model ◽  
Previous State ◽  
Chest X Ray ◽  
Tremendous Importance

Abstract Background: The Coronavirus Disease 2019 (COVID-19) is highly contagious and spreading around the world like wildfire. For this reason, fast and accurate diagnosis of COVID-19 is of tremendous importance. Chest X-rays can be used to diagnose COVID-19 but there is a global shortage of radiologists who can interpret X-rays. This study aims to develop an artificial intelligence system that can diagnose COVID-19 from chest X-rays with no human intervention.Methods: In this study, we used four pre-trained architectures - VGG16, MobileNetV2, InceptionV3 and Xception to detect COVID-19 from chest x-ray images. The architectures are optimized using Bayesian optimization and we used a weighted loss function to get better generalization capability in our models.Results: Of the four architectures we optimized, our proposed model called COVIDXception-Net performed best on the test set with an accuracy 0.94, precision 0.95, recall 0.94, and F1-score 0.94. It outperforms previous state-of-the-art methods in terms of these evaluation metrics on similar datasets. In the ablation studies, we found that the accuracy of our model dropped from 0.994 to 0.950 when we used random search instead of Bayesian optimization and 0.994 to 0.983 when we used regular loss function instead of a weighted loss function. Conclusions: In this study, we developed a model called COVIDXception-Net to diagnose COVID-19 from X-ray images. We hope that our research will assist all those involved in handling this pandemic.

Spontaneous Pneumomediastinum: Uncommon Cause of Air Leak in Late Preterm Neonate

2021 ◽  
Vol 35 (2) ◽  
pp. 93-94
Author(s):  
Jyotsna Bhushan ◽  
Shagufta Iqbal ◽  
Abhishek Chopra
Keyword(s):  
Respiratory Distress ◽  
Conservative Management ◽  
Preterm Neonate ◽  
Late Preterm ◽  
Intravenous Fluids ◽  
X Rays ◽  
Blood Gas ◽  
X Ray ◽  
Spontaneous Pneumomediastinum ◽  
Chest X Ray

A clinical case report of spontaneous pneumomediastinum in a late-preterm neonate, chest x-ray showing classical “spinnaker sail sign,” which was managed conservatively and had excellent prognosis on conservative management. Respiratory distress in a preterm neonate is a common clinical finding. Common causes include respiratory distress syndrome, transient tachypnea of the newborn, pneumonia, and pneumothorax. Pneumomediastinum is not very common cause of respiratory distress and more so spontaneous pneumomediastinum. We report here a preterm neonate with spontaneous pneumomediastinum who had excellent clinical recovery with conservative management. A male baby was delivered to G3P1A1 mother at 34 + 6 weeks through caesarean section done due to abruptio placenta. Apgar scores were 8 and 9. Maternal antenatal history was uneventful and there were no risk factors for early onset sepsis. Baby had respiratory distress soon after birth with Silverman score being 2/10. Baby was started on oxygen (O2) by nasal prongs through blender 0.5 l/min, FiO2 25%, and intravenous fluids. Blood gas done was normal. Possibility of transient tachypnea of newborn or mild hyaline membrane disease was kept. Respiratory distress increased at 20 h of life (Silverman score: 5), urgent chest x-ray done revealed “spinnaker sign” suggestive of pneumomediastinum, so baby was shifted to O2 by hood with FiO2 being 70%. Blood gas repeated was normal. Baby was managed conservatively on intravenous fluids and O2 by hood. Baby was gradually weaned off from O2 over next 5 days. As respiratory distress decreased, baby was started on orogastric feed, which baby tolerated well and then was switched to oral feeds. Serial x-rays showed resolution of pneumomediastinum. Baby was discharged on day 7 of life in stable condition on breast feeds and room air.

scholarly journals Accessory pathway analysis using a multimodal deep learning model

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Makoto Nishimori ◽  
Kunihiko Kiuchi ◽  
Kunihiro Nishimura ◽  
Kengo Kusano ◽  
Akihiro Yoshida ◽  
...  
Keyword(s):  
Deep Learning ◽  
Electrophysiological Study ◽  
Accessory Pathway ◽  
Learning Model ◽  
X Rays ◽  
X Ray ◽  
Wpw Syndrome ◽  
Chest X Ray ◽  
Deep Learning Model ◽  
Ecg Data

AbstractCardiac accessory pathways (APs) in Wolff–Parkinson–White (WPW) syndrome are conventionally diagnosed with decision tree algorithms; however, there are problems with clinical usage. We assessed the efficacy of the artificial intelligence model using electrocardiography (ECG) and chest X-rays to identify the location of APs. We retrospectively used ECG and chest X-rays to analyse 206 patients with WPW syndrome. Each AP location was defined by an electrophysiological study and divided into four classifications. We developed a deep learning model to classify AP locations and compared the accuracy with that of conventional algorithms. Moreover, 1519 chest X-ray samples from other datasets were used for prior learning, and the combined chest X-ray image and ECG data were put into the previous model to evaluate whether the accuracy improved. The convolutional neural network (CNN) model using ECG data was significantly more accurate than the conventional tree algorithm. In the multimodal model, which implemented input from the combined ECG and chest X-ray data, the accuracy was significantly improved. Deep learning with a combination of ECG and chest X-ray data could effectively identify the AP location, which may be a novel deep learning model for a multimodal model.

Sign in / Sign up

Export Citation Format

Share Document