Feasibility of an automatic computer-assisted algorithm for the detection of significant coronary artery disease in patients presenting with acute chest pain

Do plaque-related factors affect the diagnostic performance of an artificial intelligence coronary-assisted diagnosis system? Comparison with invasive coronary angiography

OBJECTIVE

The aim of this study was to investigate the effects of plaque-related factors on the diagnostic performance of an artificial intelligence coronary-assisted diagnosis system (AI-CADS).

METHODS

Patients who underwent coronary computed tomography angiography (CCTA) and invasive coronary angiography (ICA) were retrospectively included in this study. The degree of stenosis in each vessel was collected from CCTA and ICA, and the information on plaque-related factors (plaque length, plaque type, and coronary artery calcium score (CAC)) of the vessels with plaques was collected from CCTA.

RESULTS

In total, 1224 vessels in 306 patients (166 men; 65.7 ± 10.1 years) were analyzed. Of these, 391 vessels in 249 patients showed significant stenosis using ICA as the gold standard. Using per-vessel as the unit, the area under the curves of coronary stenosis ≥ 50% for AI-CADS, doctor, and AI-CADS + doctor was 0.764, 0.837, and 0.853, respectively. The accuracies in interpreting the degree of coronary stenosis were 56.0%, 68.1%, and 71.2%, respectively. Seven hundred fifty vessels showed plaques on CCTA; plaque type did not affect the interpretation results by AI-CADS (chi-square test: p = 0.0093; multiple logistic regression: p = 0.4937). However, the interpretation results for plaque length (chi-square test: p < 0.0001; multiple logistic regression: p = 0.0061) and CACs (chi-square test: p < 0.0001; multiple logistic regression: p = 0.0001) were significantly different.

CONCLUSION

AI-CADS has an ability to distinguish ≥ 50% coronary stenosis, but additional manual interpretation based on AI-CADS is necessary. The plaque length and CACs will affect the diagnostic performance of AI-CADS.

KEY POINTS

• AI-CADS can help radiologists quickly assess CCTA and improve diagnostic confidence. • Additional manual interpretation on the basis of AI-CADS is necessary. • The plaque length and CACs will affect the diagnostic performance of AI-CADS.

Root causes of adverse drug events in hospitals and artificial intelligence capabilities for prevention

AIMS

To identify and prioritize the root causes of adverse drug events (ADEs) in hospitals and to assess the ability of artificial intelligence (AI) capabilities to prevent ADEs.

DESIGN

A mixed method design was used.

METHODS

A cross-sectional study for hospitals in Spain was carried out between February and April 2019 to identify and prioritize the root causes of ADEs. A nominal group technique was also used to assess the ability of AI capabilities to prevent ADEs.

RESULTS

The main root cause of ADEs was a lack of adherence to safety protocols (64.8%), followed by identification errors (57.4%), and fragile and polymedicated patients (44.4%). An analysis of the AI capabilities to prevent the root causes of ADEs showed that identification and reading are two potentially useful capabilities.

CONCLUSION

Identification error is one of the main root causes of drug adverse events and AI capabilities could potentially prevent drug adverse events.

IMPACT

This study highlights the role of AI capabilities in safely identifying both patients and drugs, which is a crucial part of the medication administration process, and how this can prevent ADEs in hospitals.

Accuracy of an Artificial Intelligence Deep Learning Algorithm Implementing a Recurrent Neural Network With Long Short-term Memory for the Automated Detection of Calcified Plaques From Coronary Computed Tomography Angiography

PURPOSE

The purpose of this study was to evaluate the accuracy of a novel fully automated deep learning (DL) algorithm implementing a recurrent neural network (RNN) with long short-term memory (LSTM) for the detection of coronary artery calcium (CAC) from coronary computed tomography angiography (CCTA) data.

MATERIALS AND METHODS

Under an IRB waiver and in HIPAA compliance, a total of 194 patients who had undergone CCTA were retrospectively included. Two observers independently evaluated the image quality and recorded the presence of CAC in the right (RCA), the combination of left main and left anterior descending (LM-LAD), and left circumflex (LCx) coronary arteries. Noncontrast CACS scans were allowed to be used in cases of uncertainty. Heart and coronary artery centerline detection and labeling were automatically performed. Presence of CAC was assessed by a RNN-LSTM. The algorithm's overall and per-vessel sensitivity, specificity, and diagnostic accuracy were calculated.

RESULTS

CAC was absent in 84 and present in 110 patients. As regards CCTA, the median subjective image quality, signal-to-noise ratio, and contrast-to-noise ratio were 3.0, 13.0, and 11.4. A total of 565 vessels were evaluated. On a per-vessel basis, the algorithm achieved a sensitivity, specificity, and diagnostic accuracy of 93.1% (confidence interval [CI], 84.3%-96.7%), 82.76% (CI, 74.6%-89.4%), and 86.7% (CI, 76.8%-87.9%), respectively, for the RCA, 93.1% (CI, 86.4%-97.7%), 95.5% (CI, 88.77%-98.75%), and 94.2% (CI. 90.2%-94.6%), respectively, for the LM-LAD, and 89.9% (CI, 80.2%-95.8%), 90.0% (CI, 83.2%-94.7%), and 89.9% (CI, 85.0%-94.1%), respectively, for the LCx. The overall sensitivity, specificity, and diagnostic accuracy were 92.1% (CI, 92.1%-95.2%), 88.9% (CI. 84.9%-92.1%), and 90.3% (CI, 88.0%-90.0%), respectively. When accounting for image quality, the algorithm achieved a sensitivity, specificity, and diagnostic accuracy of 76.2%, 87.5%, and 82.2%, respectively, for poor-quality data sets and 93.3%, 89.2% and 90.9%, respectively, when data sets rated adequate or higher were combined.

CONCLUSION

The proposed RNN-LSTM demonstrated high diagnostic accuracy for the detection of CAC from CCTA.

Computer-aided detection of bone metastasis in bone scintigraphy images using parallelepiped classification method

OBJECTIVE

Accurate diagnosis of metastatic tissue on bone scintigraphy images is of paramount importance in making treatment decisions. Although several automated systems have developed, more and better interpretation methods are still being sought. In the present study, a new modality for bone metastasis detection from bone scintigraphy images using parallelepiped classification (PC) as method for mapping the radionuclide distribution is presented.

METHODS

Bone scintigraphy images from 12 patients with bone metastases were analyzed using the parallelepiped classifier that generated color maps of scintigraphic images. Seven classes of radionuclide accumulation have been identified and fed into machine learning software. The accuracy of the proposed method was evaluated by statistical measurements in a confusion matrix. Overall accuracy, producer's and user's accuracies and κ coefficient were computed from each confusion matrix associated with the individual case.

RESULTS

The results revealed that the method is sufficiently precise to differentiate the metastatic bone from normal tissue (overall classification accuracy = 87.58 ± 2.25% and κ coefficient = 0.8367 ± 0.0252). The maps are easier to read (due to better contrast) and can detect even slightest differences in accumulation levels among pixels.

CONCLUSIONS

In conclusion, these preliminary data suggest that bone scintigraphy combined with PC method could play an important role in the detection of bone metastasis, allowing for an easier but correct interpretation of the images, with effects on the diagnosis accuracy and decision making on the treatment to be applied.

A comparison of computer-assisted detection (CAD) programs for the identification of colorectal polyps: performance and sensitivity analysis, current limitations and practical tips for radiologists

AIM

To directly compare the accuracy and speed of analysis of two commercially available computer-assisted detection (CAD) programs in detecting colorectal polyps.

MATERIALS AND METHOD

In this retrospective single-centre study, patients who had colorectal polyps identified on computed tomography colonography (CTC) and subsequent lower gastrointestinal endoscopy, were analysed using two commercially available CAD programs (CAD1 and CAD2). Results were compared against endoscopy to ascertain sensitivity and positive predictive value (PPV) for colorectal polyps. Time taken for CAD analysis was also calculated.

RESULTS

CAD1 demonstrated a sensitivity of 89.8%, PPV of 17.6% and mean analysis time of 125.8 seconds. CAD2 demonstrated a sensitivity of 75.5%, PPV of 44.0% and mean analysis time of 84.6 seconds.

CONCLUSION

The sensitivity and PPV for colorectal polyps and CAD analysis times can vary widely between current commercially available CAD programs. There is still room for improvement. Generally, there is a trade-off between sensitivity and PPV, and so further developments should aim to optimise both. Information on these factors should be made routinely available, so that an informed choice on their use can be made. This information could also potentially influence the radiologist's use of CAD results.

Automated Computer-Assisted Diagnosis of Obstructive Coronary Artery Disease in Emergency Department Patients Undergoing 256-Slice Coronary Computed Tomography Angiography for Acute Chest Pain

A 256-slice coronary computed tomography angiography (CCTA) is an accurate method for detection and exclusion of obstructive coronary artery disease (OBS-CAD). However, accurate image interpretation requires expertise and may not be available at all hours. The purpose of this study was to evaluate the usefulness of a fully automated computer-assisted diagnosis (COMP-DIAG) tool for exclusion of OBS-CAD in patients in the emergency department (ED) presenting with chest pain. Three hundred sixty-nine patients in ED without known coronary disease underwent 256-slice CCTA as part of the assessment of chest pain of uncertain origin. COMP-DIAG (CorAnalyzer II) automatically reported presence or exclusion of OBS-CAD (>50% stenosis, ≥1 vessel). Performance characteristics of COMP-DIAG for exclusion and detection of OBS-CAD were determined using expert reading as the reference standard. Seventeen (5%) studies were unassessable by COMP-DIAG software, and 352 patients (1,056 vessels) were therefore available for analysis. COMP-DIAG identified 33% of assessable studies as having OBS-CAD, but the prevalence of OBS-CAD on CCTA was only 18% (66 of 352 patients) by standard expert reading. However, COMP-DIAG correctly identified 61 of the 66 patients (93%) with OBS-CAD with 21 vessels (2%) with OBS-CAD misclassified as negative. In conclusion, compared to expert reading, automated computer-assisted diagnosis using the CorAnalyzer showed high sensitivity but only moderate specificity for detection of obstructive coronary disease in patients in ED who underwent 256-slice CCTA. The high negative predictive value of this computer-assisted algorithm may be useful in the ED setting.

Computer-aided analysis of 64-slice coronary computed tomography angiography: a comparison with manual interpretation

Coronary computed tomography angiography (CCTA) is increasingly used for the assessment of coronary heart disease (CHD) in symptomatic patients. Software applications have recently been developed to facilitate efficient and accurate analysis of CCTA. This study aims to evaluate the clinical application of computer-aided diagnosis (CAD) software for the detection of significant coronary stenosis on CCTA in populations with low (8%), moderate (13%), and high (27%) CHD prevalence. A total of 341 consecutive patients underwent 64-slice CCTA at 3 clinical sites in the United States. CAD software performed automatic detection of significant coronary lesions (>50% stenosis). CAD results were then compared to the consensus manual interpretation of 2 imaging experts. Data analysis was conducted for each patient and segment. The CAD had 100% sensitivity per patient across all 3 clinical sites. Specificity in the low, moderate, and high CHD prevalence populations was 64%, 41%, and 38%, respectively. The negative predictive value at the 3 clinical sites was 100%. The positive predictive value was 22%, 21%, and 38% for the low, moderate, and high CHD prevalence populations, respectively. This study demonstrates the utility of CAD software in 3 distinct clinical settings. In a low-prevalence population, such as seen in the emergency department, CAD can be used as a Computer-Aided Simple Triage tool to assist in diagnostic delineation of acute chest pain. In a higher prevalence population, CAD software is useful as an adjunct for both the experienced and inexperienced reader.

Computer-aided simple triage (CAST) for coronary CT angiography (CCTA)

PURPOSE

Following a recent introduction of computer-aided simple triage (CAST) as a new subclass of computer-aided detection/diagnosis (CAD), we present a CAST software system for a fully automatic initial interpretation of coronary CT angiography (CCTA). We show how the system design and diagnostic performance make it CAST-compliant and suitable for chest pain patient triage in emergency room (ER).

METHODS

The processing performed by the system consists of three major steps: segmentation of coronary artery tree, labeling of major coronary arteries, and detection of significant stenotic lesions (causing > 50% stenosis). In addition, the system performs an automatic image quality assessment to discards low-quality studies. For multiphase studies, the system automatically chooses the best phase for each coronary artery. Clinical evaluation results were collected in 14 independent trials that included more than 2000 CCTA studies. Automatic diagnosis results were compared with human interpretation of the CCTA and to cath lab results.

RESULTS

The presented system performs a fully automatic initial interpretation of CCTA without any human interaction and detects studies with significant coronary artery disease. The system demonstrated higher than 90% per patient sensitivity and 40-70% per patient specificity. For the chest pain, ER population, the specificity was 60-70%, yielding higher than 98% NPV.

CONCLUSIONS

The diagnostic performance of the presented CCTA CAD system meets the CAST requirements, thus enabling efficient, 24/7 utilization of CCTA for chest pain patient triage in ER. This is the first fully operational, clinically validated, CAST-compliant CAD system for a fully automatic analysis of CCTA and detection of significant stenosis.