Automatized Detection and Categorization of Fissure Sealants from Intraoral Digital Photographs Using Artificial Intelligence

The aim of the present study was to investigate the diagnostic performance of a trained convolutional neural network (CNN) for detecting and categorizing fissure sealants from intraoral photographs using the expert standard as reference. An image set consisting of 2352 digital photographs from permanent posterior teeth (461 unsealed tooth surfaces/1891 sealed surfaces) was divided into a training set (n = 1881/364/1517) and a test set (n = 471/97/374). All the images were scored according to the following categories: unsealed molar, intact, sufficient and insufficient sealant. Expert diagnoses served as the reference standard for cyclic training and repeated evaluation of the CNN (ResNeXt-101-32x8d), which was trained by using image augmentation and transfer learning. A statistical analysis was performed, including the calculation of contingency tables and areas under the receiver operating characteristic curve (AUC). The results showed that the CNN accurately detected sealants in 98.7% of all the test images, corresponding to an AUC of 0.996. The diagnostic accuracy and AUC were 89.6% and 0.951, respectively, for intact sealant; 83.2% and 0.888, respectively, for sufficient sealant; 92.4 and 0.942, respectively, for insufficient sealant. On the basis of the documented results, it was concluded that good agreement with the reference standard could be achieved for automatized sealant detection by using artificial intelligence methods. Nevertheless, further research is necessary to improve the model performance.

Download Full-text

Caries Detection on Intraoral Images Using Artificial Intelligence

Journal of Dental Research ◽

10.1177/00220345211032524 ◽

2021 ◽

pp. 002203452110325

Author(s):

J. Kühnisch ◽

O. Meyer ◽

M. Hesenius ◽

R. Hickel ◽

V. Gruhn

Keyword(s):

Artificial Intelligence ◽

Current Approach ◽

Test Method ◽

Automated Image Analysis ◽

Caries Detection ◽

Permanent Teeth ◽

Diagnostic Study ◽

Tooth Surfaces ◽

Image Set ◽

Machine Readable

Although visual examination (VE) is the preferred method for caries detection, the analysis of intraoral digital photographs in machine-readable form can be considered equivalent to VE. While photographic images are rarely used in clinical practice for diagnostic purposes, they are the fundamental requirement for automated image analysis when using artificial intelligence (AI) methods. Considering that AI has not been used for automatic caries detection on intraoral images so far, this diagnostic study aimed to develop a deep learning approach with convolutional neural networks (CNNs) for caries detection and categorization (test method) and to compare the diagnostic performance with respect to expert standards. The study material consisted of 2,417 anonymized photographs from permanent teeth with 1,317 occlusal and 1,100 smooth surfaces. All the images were evaluated into the following categories: caries free, noncavitated caries lesion, or caries-related cavitation. Each expert diagnosis served as a reference standard for cyclic training and repeated evaluation of the AI methods. The CNN was trained using image augmentation and transfer learning. Before training, the entire image set was divided into a training and test set. Validation was conducted by selecting 25%, 50%, 75%, and 100% of the available images from the training set. The statistical analysis included calculations of the sensitivity (SE), specificity (SP), and area under the receiver operating characteristic (ROC) curve (AUC). The CNN was able to correctly detect caries in 92.5% of cases when all test images were considered (SE, 89.6; SP, 94.3; AUC, 0.964). If the threshold of caries-related cavitation was chosen, 93.3% of all tooth surfaces were correctly classified (SE, 95.7; SP, 81.5; AUC, 0.955). It can be concluded that it was possible to achieve more than 90% agreement in caries detection using the AI method with standardized, single-tooth photographs. Nevertheless, the current approach needs further improvement.

Download Full-text

Chest radiograph-based artificial intelligence predictive model for mortality in community-acquired pneumonia

BMJ Open Respiratory Research ◽

10.1136/bmjresp-2021-001045 ◽

2021 ◽

Vol 8 (1) ◽

pp. e001045

Author(s):

Jessica Quah ◽

Charlene Jin Yee Liew ◽

Lin Zou ◽

Xuan Han Koh ◽

Rayan Alsuwaigh ◽

...

Keyword(s):

Artificial Intelligence ◽

Chest Radiograph ◽

Risk Score ◽

Mortality Risk ◽

Characteristic Curve ◽

Pneumonia Severity Index ◽

Model Performance ◽

Community Acquired Pneumonia ◽

Pneumonia Severity ◽

Severity Scores

BackgroundChest radiograph (CXR) is a basic diagnostic test in community-acquired pneumonia (CAP) with prognostic value. We developed a CXR-based artificial intelligence (AI) model (CAP AI predictive Engine: CAPE) and prospectively evaluated its discrimination for 30-day mortality.MethodsDeep-learning model using convolutional neural network (CNN) was trained with a retrospective cohort of 2235 CXRs from 1966 unique adult patients admitted for CAP from 1 January 2019 to 31 December 2019. A single-centre prospective cohort between 11 May 2020 and 15 June 2020 was analysed for model performance. CAPE mortality risk score based on CNN analysis of the first CXR performed for CAP was used to determine the area under the receiver operating characteristic curve (AUC) for 30-day mortality.Results315 inpatient episodes for CAP occurred, with 30-day mortality of 19.4% (n=61/315). Non-survivors were older than survivors (mean (SD)age, 80.4 (10.3) vs 69.2 (18.7)); more likely to have dementia (n=27/61 vs n=58/254) and malignancies (n=16/61 vs n=18/254); demonstrate higher serum C reactive protein (mean (SD), 109 mg/L (98.6) vs 59.3 mg/L (69.7)) and serum procalcitonin (mean (SD), 11.3 (27.8) μg/L vs 1.4 (5.9) μg/L). The AUC for CAPE mortality risk score for 30-day mortality was 0.79 (95% CI 0.73 to 0.85, p<0.001); Pneumonia Severity Index (PSI) 0.80 (95% CI 0.74 to 0.86, p<0.001); Confusion of new onset, blood Urea nitrogen, Respiratory rate, Blood pressure, 65 (CURB-65) score 0.76 (95% CI 0.70 to 0.81, p<0.001), respectively. CAPE combined with CURB-65 model has an AUC of 0.83 (95% CI 0.77 to 0.88, p<0.001). The best performing model was CAPE incorporated with PSI, with an AUC of 0.84 (95% CI 0.79 to 0.89, p<0.001).ConclusionCXR-based CAPE mortality risk score was comparable to traditional pneumonia severity scores and improved its discrimination when combined.

Download Full-text

ICD10Net: An Artificial Intelligence Algorithm with Medical Background Conducts ICD-10-CM Coding Task with Outstanding Performance (Preprint)

10.2196/preprints.13677 ◽

2019 ◽

Author(s):

Chin Lin ◽

Yu-Sheng Lou ◽

Chia-Cheng Lee ◽

Chia-Jung Hsu ◽

Ding-Chung Wu ◽

...

Keyword(s):

Artificial Intelligence ◽

General Hospital ◽

Pearson Correlation ◽

Model Performance ◽

International Classification Of Diseases ◽

Free Text ◽

Daily Work ◽

Medical Background ◽

Icd 10 ◽

F Measure

BACKGROUND An artificial intelligence-based algorithm has shown a powerful ability for coding the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) in discharge notes. However, its performance still requires improvement compared with human experts. The major disadvantage of the previous algorithm is its lack of understanding medical terminologies. OBJECTIVE We propose some methods based on human-learning process and conduct a series of experiments to validate their improvements. METHODS We compared two data sources for training the word-embedding model: English Wikipedia and PubMed journal abstracts. Moreover, the fixed, changeable, and double-channel embedding tables were used to test their performance. Some additional tricks were also applied to improve accuracy. We used these methods to identify the three-chapter-level ICD-10-CM diagnosis codes in a set of discharge notes. Subsequently, 94,483-labeled discharge notes from June 1, 2015 to June 30, 2017 were used from the Tri-Service General Hospital in Taipei, Taiwan. To evaluate performance, 24,762 discharge notes from July 1, 2017 to December 31, 2017, from the same hospital were used. Moreover, 74,324 additional discharge notes collected from other seven hospitals were also tested. The F-measure is the major global measure of effectiveness. RESULTS In understanding medical terminologies, the PubMed-embedding model (Pearson correlation = 0.60/0.57) shows a better performance compared with the Wikipedia-embedding model (Pearson correlation = 0.35/0.31). In the accuracy of ICD-10-CM coding, the changeable model both used the PubMed- and Wikipedia-embedding model has the highest testing mean F-measure (0.7311 and 0.6639 in Tri-Service General Hospital and other seven hospitals, respectively). Moreover, a proposed method called a hybrid sampling method, an augmentation trick to avoid algorithms identifying negative terms, was found to additionally improve the model performance. CONCLUSIONS The proposed model architecture and training method is named as ICD10Net, which is the first expert level model practically applied to daily work. This model can also be applied in unstructured information extraction from free-text medical writing. We have developed a web app to demonstrate our work (https://linchin.ndmctsgh.edu.tw/app/ICD10/).

Download Full-text

Artificial Intelligence for Skin Cancer Detection: A Scoping Review (Preprint)

10.2196/preprints.22934 ◽

2020 ◽

Author(s):

Abdulrahman Takiddin ◽

Jens Schneider ◽

Yin Yang ◽

Alaa Abd-Alrazaq ◽

Mowafa Househ

Keyword(s):

Artificial Intelligence ◽

Skin Cancer ◽

Scoping Review ◽

Performance Metrics ◽

Data Extraction ◽

Characteristic Curve ◽

Evaluation Metrics ◽

Cancer Type ◽

Computer Algorithms ◽

Scoping Reviews

BACKGROUND Skin cancer is the most common cancer type affecting humans. Traditional skin cancer diagnosis methods are costly, require a professional physician, and take time. Hence, to aid in diagnosing skin cancer, Artificial Intelligence (AI) tools are being used, including shallow and deep machine learning-based techniques that are trained to detect and classify skin cancer using computer algorithms and deep neural networks. OBJECTIVE The aim of this study is to identify and group the different types of AI-based technologies used to detect and classify skin cancer. The study also examines the reliability of the selected papers by studying the correlation between the dataset size and number of diagnostic classes with the performance metrics used to evaluate the models. METHODS We conducted a systematic search for articles using IEEE Xplore, ACM DL, and Ovid MEDLINE databases following the PRISMA Extension for Scoping Reviews (PRISMA-ScR) guidelines. The study included in this scoping review had to fulfill several selection criteria; to be specifically about skin cancer, detecting or classifying skin cancer, and using AI technologies. Study selection and data extraction were conducted by two reviewers independently. Extracted data were synthesized narratively, where studies were grouped based on the diagnostic AI techniques and their evaluation metrics. RESULTS We retrieved 906 papers from the 3 databases, but 53 studies were eligible for this review. While shallow techniques were used in 14 studies, deep techniques were utilized in 39 studies. The studies used accuracy (n=43/53), the area under receiver operating characteristic curve (n=5/53), sensitivity (n=3/53), and F1-score (n=2/53) to assess the proposed models. Studies that use smaller datasets and fewer diagnostic classes tend to have higher reported accuracy scores. CONCLUSIONS The adaptation of AI in the medical field facilitates the diagnosis process of skin cancer. However, the reliability of most AI tools is questionable since small datasets or low numbers of diagnostic classes are used. In addition, a direct comparison between methods is hindered by a varied use of different evaluation metrics and image types.

Download Full-text

Design of English reading and learning management system in college education based on artificial intelligence

Journal of Intelligent & Fuzzy Systems ◽

10.3233/jifs-219125 ◽

2021 ◽

pp. 1-10

Author(s):

Fen Zhang ◽

Min She

Keyword(s):

Artificial Intelligence ◽

Model Performance ◽

College Education ◽

Personalized Learning ◽

Learning System ◽

Model Function ◽

English Reading ◽

Function Modules ◽

Learning Platforms ◽

Learning Management

English reading learning in college education is an efficient means of English learning. However, most of the current English reading learning platforms in colleges and universities only put different English books on the platform in electronic form for students to read, which leads to blindness of reading. Based on artificial intelligence algorithms, this paper builds model function modules according to the needs of English reading and learning management in college education and implements system functions based on artificial intelligence algorithms. Moreover, according to the above design principles of personalized learning model and the characteristics of personalized network learning, this paper designs a personalized learning system based on meaningful learning theory. In addition, this article verifies and analyzes the model performance. The research results show that the model proposed in this paper has a certain effect.

Download Full-text

Stress testing reveals gaps in clinic readiness of image-based diagnostic artificial intelligence models

npj Digital Medicine ◽

10.1038/s41746-020-00380-6 ◽

2021 ◽

Vol 4 (1) ◽

Author(s):

Albert T. Young ◽

Kristen Fernandez ◽

Jacob Pfau ◽

Rasika Reddy ◽

Nhat Anh Cao ◽

...

Keyword(s):

Artificial Intelligence ◽

Real World ◽

Stress Testing ◽

Operating Characteristic ◽

Characteristic Curve ◽

Skin Lesions ◽

Stress Tests ◽

Performance Metric ◽

Intelligence Models ◽

Artificial Intelligence Models

AbstractArtificial intelligence models match or exceed dermatologists in melanoma image classification. Less is known about their robustness against real-world variations, and clinicians may incorrectly assume that a model with an acceptable area under the receiver operating characteristic curve or related performance metric is ready for clinical use. Here, we systematically assessed the performance of dermatologist-level convolutional neural networks (CNNs) on real-world non-curated images by applying computational “stress tests”. Our goal was to create a proxy environment in which to comprehensively test the generalizability of off-the-shelf CNNs developed without training or evaluation protocols specific to individual clinics. We found inconsistent predictions on images captured repeatedly in the same setting or subjected to simple transformations (e.g., rotation). Such transformations resulted in false positive or negative predictions for 6.5–22% of skin lesions across test datasets. Our findings indicate that models meeting conventionally reported metrics need further validation with computational stress tests to assess clinic readiness.

Download Full-text

Deep Learning Versus Logistic Regression to Predict Opioid Dose Reduction After Spinal Cord Stimulation

Neurosurgery ◽

10.1093/neuros/nyaa447_638 ◽

2020 ◽

Vol 67 (Supplement_1) ◽

Author(s):

Syed M Adil ◽

Lefko T Charalambous ◽

Kelly R Murphy ◽

Shervin Rahimpour ◽

Stephen C Harward ◽

...

Keyword(s):

Neural Network ◽

Spinal Cord ◽

Deep Learning ◽

Dose Reduction ◽

Spinal Cord Stimulation ◽

Network Architecture ◽

Characteristic Curve ◽

Model Performance ◽

Health Crisis ◽

Opioid Dose

Abstract INTRODUCTION Opioid misuse persists as a public health crisis affecting approximately one in four Americans.1 Spinal cord stimulation (SCS) is a neuromodulation strategy to treat chronic pain, with one goal being decreased opioid consumption. Accurate prognostication about SCS success is key in optimizing surgical decision making for both physicians and patients. Deep learning, using neural network models such as the multilayer perceptron (MLP), enables accurate prediction of non-linear patterns and has widespread applications in healthcare. METHODS The IBM MarketScan® (IBM) database was queried for all patients ≥ 18 years old undergoing SCS from January 2010 to December 2015. Patients were categorized into opioid dose groups as follows: No Use, ≤ 20 morphine milligram equivalents (MME), 20–50 MME, 50–90 MME, and >90 MME. We defined “opiate weaning” as moving into a lower opioid dose group (or remaining in the No Use group) during the 12 months following permanent SCS implantation. After pre-processing, there were 62 predictors spanning demographics, comorbidities, and pain medication history. We compared an MLP with four hidden layers to the LR model with L1 regularization. Model performance was assessed using area under the receiver operating characteristic curve (AUC) with 5-fold nested cross-validation. RESULTS Ultimately, 6,124 patients were included, of which 77% had used opioids for >90 days within the 1-year pre-SCS and 72% had used >5 types of medications during the 90 days prior to SCS. The mean age was 56 ± 13 years old. Collectively, 2,037 (33%) patients experienced opiate weaning. The AUC was 0.74 for the MLP and 0.73 for the LR model. CONCLUSION To our knowledge, we present the first use of deep learning to predict opioid weaning after SCS. Model performance was slightly better than regularized LR. Future efforts should focus on optimization of neural network architecture and hyperparameters to further improve model performance. Models should also be calibrated and externally validated on an independent dataset. Ultimately, such tools may assist both physicians and patients in predicting opioid dose reduction after SCS.

Download Full-text

Artificial Intelligence Coreflooding Simulator for Special Core Data Analysis

SPE Reservoir Evaluation & Engineering ◽

10.2118/202700-pa ◽

2021 ◽

pp. 1-29

Author(s):

Eric Sonny Mathew ◽

Moussa Tembely ◽

Waleed AlAmeri ◽

Emad W. Al-Shalabi ◽

Abdul Ravoof Shaik

Keyword(s):

Artificial Intelligence ◽

Steady State ◽

History Matching ◽

Water Saturation ◽

Flow Behavior ◽

Model Performance ◽

Analysis Data ◽

Coefficient Of Determination ◽

Water Drainage ◽

Data Set

Two of the most critical properties for multiphase flow in a reservoir are relative permeability (Kr) and capillary pressure (Pc). To determine these parameters, careful interpretation of coreflooding and centrifuge experiments is necessary. In this work, a machine learning (ML) technique was incorporated to assist in the determination of these parameters quickly and synchronously for steady-state drainage coreflooding experiments. A state-of-the-art framework was developed in which a large database of Kr and Pc curves was generated based on existing mathematical models. This database was used to perform thousands of coreflood simulation runs representing oil-water drainage steady-state experiments. The results obtained from the corefloods including pressure drop and water saturation profile, along with other conventional core analysis data, were fed as features into the ML model. The entire data set was split into 70% for training, 15% for validation, and the remaining 15% for the blind testing of the model. The 70% of the data set for training teaches the model to capture fluid flow behavior inside the core, and then 15% of the data set was used to validate the trained model and to optimize the hyperparameters of the ML algorithm. The remaining 15% of the data set was used for testing the model and assessing the model performance scores. In addition, K-fold split technique was used to split the 15% testing data set to provide an unbiased estimate of the final model performance. The trained/tested model was thereby used to estimate Kr and Pc curves based on available experimental results. The values of the coefficient of determination (R2) were used to assess the accuracy and efficiency of the developed model. The respective crossplots indicate that the model is capable of making accurate predictions with an error percentage of less than 2% on history matching experimental data. This implies that the artificial-intelligence- (AI-) based model is capable of determining Kr and Pc curves. The present work could be an alternative approach to existing methods for interpreting Kr and Pc curves. In addition, the ML model can be adapted to produce results that include multiple options for Kr and Pc curves from which the best solution can be determined using engineering judgment. This is unlike solutions from some of the existing commercial codes, which usually provide only a single solution. The model currently focuses on the prediction of Kr and Pc curves for drainage steady-state experiments; however, the work can be extended to capture the imbibition cycle as well.

Download Full-text

Artificial Intelligence Is a Promising Prospect for the Detection of Prostate Cancer Extracapsular Extension With Mp-mri: A Two-center Comparative Study

10.21203/rs.3.rs-298296/v1 ◽

2021 ◽

Author(s):

Ying Hou ◽

Yi-Hong Zhang ◽

Jie Bao ◽

Mei-Ling Bao ◽

Guang Yang ◽

...

Keyword(s):

Prostate Cancer ◽

Artificial Intelligence ◽

Paired Comparison ◽

Characteristic Curve ◽

Multiparametric Mri ◽

Training Data ◽

Urinary Continence ◽

Data Sets ◽

Extracapsular Extension ◽

Promising Alternative

Abstract Purpose: A balance between preserving urinary continence and achievement of negative margins is of clinical relevance while implementary difficulty. Preoperatively accurate detection of prostate cancer (PCa) extracapsular extension (ECE) is thus crucial for determining appropriate treatment options. We aimed to develop and clinically validate an artificial intelligence (AI)-assisted tool for the detection of ECE in patients with PCa using multiparametric MRI. Methods: 849 patients with localized PCa underwent multiparametric MRI before radical prostatectomy were retrospectively included from two medical centers. The AI tool was built on a ResNeXt network embedded with a spatial attention map of experts’ prior knowledges (PAGNet) from 596 training data sets. The tool was validated in 150 internal and 103 external data sets, respectively; and its clinical applicability was compared with expert-based interpretation and AI-expert interaction.Results: An index PAGNet model using a single-slice image yielded the highest areas under the receiver operating characteristic curve (AUC) of 0.857 (95% confidence interval [CI], 0.827-0.884), 0.807 (95% CI, 0.735-0.867) and 0.728 (95% CI, 0.631-0.811) in the training, internal test and external test cohorts, compared to the conventional ResNeXt networks. For experts, the inter-reader agreement was observed in only 437/849 (51.5%) patients with a Kappa value 0.343. And the performance of two experts (AUC, 0.632 to 0.741 vs 0.715 to 0.857) was lower (paired comparison, all p values < 0.05) than that of AI assessment. When expert’ interpretations were adjusted by the AI assessments, the performance of both two experts was improved.Conclusion: Our AI tool, showing improved accuracy, offers a promising alternative to human experts for imaging staging of PCa ECE using multiparametric MRI.

Download Full-text

ECG-AI: electrocardiographic artificial intelligence model for prediction of heart failure

European Heart Journal - Digital Health ◽

10.1093/ehjdh/ztab080 ◽

2021 ◽

Author(s):

Oguz Akbilgic ◽

Liam Butler ◽

Ibrahim Karabayir ◽

Patricia P Chang ◽

Dalane W Kitzman ◽

...

Keyword(s):

Neural Network ◽

Artificial Intelligence ◽

Heart Failure ◽

Convolutional Neural Network ◽

Prediction Models ◽

Smoking Status ◽

Characteristic Curve ◽

Gradient Boosting ◽

Model Output ◽

Light Gradient

Abstract Aims Heart failure (HF) is a leading cause of death. Early intervention is the key to reduce HF-related morbidity and mortality. This study assesses the utility of electrocardiograms (ECGs) in HF risk prediction. Methods and results Data from the baseline visits (1987–89) of the Atherosclerosis Risk in Communities (ARIC) study was used. Incident hospitalized HF events were ascertained by ICD codes. Participants with good quality baseline ECGs were included. Participants with prevalent HF were excluded. ECG-artificial intelligence (AI) model to predict HF was created as a deep residual convolutional neural network (CNN) utilizing standard 12-lead ECG. The area under the receiver operating characteristic curve (AUC) was used to evaluate prediction models including (CNN), light gradient boosting machines (LGBM), and Cox proportional hazards regression. A total of 14 613 (45% male, 73% of white, mean age ± standard deviation of 54 ± 5) participants were eligible. A total of 803 (5.5%) participants developed HF within 10 years from baseline. Convolutional neural network utilizing solely ECG achieved an AUC of 0.756 (0.717–0.795) on the hold-out test data. ARIC and Framingham Heart Study (FHS) HF risk calculators yielded AUC of 0.802 (0.750–0.850) and 0.780 (0.740–0.830). The highest AUC of 0.818 (0.778–0.859) was obtained when ECG-AI model output, age, gender, race, body mass index, smoking status, prevalent coronary heart disease, diabetes mellitus, systolic blood pressure, and heart rate were used as predictors of HF within LGBM. The ECG-AI model output was the most important predictor of HF. Conclusions ECG-AI model based solely on information extracted from ECG independently predicts HF with accuracy comparable to existing FHS and ARIC risk calculators.

Download Full-text