Computed tomography as a first-line investigation for elderly patients admitted to a surgical assessment unit

A study on whether deep learning models based on CT images for bone density classification and prediction can be used for opportunistic osteoporosis screening

This study utilized deep learning to classify osteoporosis and predict bone density using opportunistic CT scans and independently tested the models on data from different hospitals and equipment. Results showed high accuracy and strong correlation with QCT results, showing promise for expanding osteoporosis screening and reducing unnecessary radiation and costs.

PURPOSE

To explore the feasibility of using deep learning to establish a model for osteoporosis classification and bone density value prediction based on opportunistic CT scans and to verify its generalization and diagnostic ability using an independent test set.

METHODS

A total of 1219 cases of opportunistic CT scans were included in this study, with QCT results as the reference standard. The training set: test set: independent test set ratio was 703: 176: 340, and the independent test set data of 340 cases were from 3 different hospitals and 4 different CT scanners. The VB-Net structure automatic segmentation model was used to segment the trabecular bone, and DenseNet was used to establish a three-classification model and bone density value prediction regression model. The performance parameters of the models were calculated and evaluated.

RESULTS

The ROC curves showed that the mean AUCs of the three-category classification model for categorizing cases into "normal," "osteopenia," and "osteoporosis" for the training set, test set, and independent test set were 0.999, 0.970, and 0.933, respectively. The F1 score, accuracy, precision, recall, precision, and specificity of the test set were 0.903, 0.909, 0.899, 0.908, and 0.956, respectively, and those of the independent test set were 0.798, 0.815, 0.792, 0.81, and 0.899, respectively. The MAEs of the bone density prediction regression model in the training set, test set, and independent test set were 3.15, 6.303, and 10.257, respectively, and the RMSEs were 4.127, 8.561, and 13.507, respectively. The R-squared values were 0.991, 0.962, and 0.878, respectively. The Pearson correlation coefficients were 0.996, 0.981, and 0.94, respectively, and the p values were all < 0.001. The predicted values and bone density values were highly positively correlated, and there was a significant linear relationship.

CONCLUSION

Using deep learning neural networks to process opportunistic CT scan images of the body can accurately predict bone density values and perform bone density three-classification diagnosis, which can reduce the radiation risk, economic consumption, and time consumption brought by specialized bone density measurement, expand the scope of osteoporosis screening, and have broad application prospects.

Automated detection of the contrast phase in MDCT by an artificial neural network improves the accuracy of opportunistic bone mineral density measurements

Objectives

To determine the accuracy of an artificial neural network (ANN) for fully automated detection of the presence and phase of iodinated contrast agent in routine abdominal multidetector computed tomography (MDCT) scans and evaluate the effect of contrast correction for osteoporosis screening.

Methods

This HIPPA-compliant study retrospectively included 579 MDCT scans in 193 patients (62.4 ± 14.6 years, 48 women). Three different ANN models (2D DenseNet with random slice selection, 2D DenseNet with anatomy-guided slice selection, 3D DenseNet) were trained in 462 MDCT scans of 154 patients (threefold cross-validation), who underwent triphasic CT. All ANN models were tested in 117 unseen triphasic scans of 39 patients, as well as in a public MDCT dataset containing 311 patients. In the triphasic test scans, trabecular volumetric bone mineral density (BMD) was calculated using a fully automated pipeline. Root-mean-square errors (RMSE) of BMD measurements with and without correction for contrast application were calculated in comparison to nonenhanced (NE) scans.

Results

The 2D DenseNet with anatomy-guided slice selection outperformed the competing models and achieved an F1 score of 0.98 and an accuracy of 98.3% in the test set (public dataset: F1 score 0.93; accuracy 94.2%). Application of contrast agent resulted in significant BMD biases (all p < .001; portal-venous (PV): RMSE 18.7 mg/ml, mean difference 17.5 mg/ml; arterial (AR): RMSE 6.92 mg/ml, mean difference 5.68 mg/ml). After the fully automated correction, this bias was no longer significant (p > .05; PV: RMSE 9.45 mg/ml, mean difference 1.28 mg/ml; AR: RMSE 3.98 mg/ml, mean difference 0.94 mg/ml).

Conclusion

Automatic detection of the contrast phase in multicenter CT data was achieved with high accuracy, minimizing the contrast-induced error in BMD measurements.

Key Points

• A 2D DenseNet with anatomy-guided slice selection achieved an F1 score of 0.98 and an accuracy of 98.3% in the test set. In a public dataset, an F1 score of 0.93 and an accuracy of 94.2% were obtained.

• Automated adjustment for contrast injection improved the accuracy of lumbar bone mineral density measurements (RMSE 18.7 mg/ml vs. 9.45 mg/ml respectively, in the portal-venous phase).

• An artificial neural network can reliably reveal the presence and phase of iodinated contrast agent in multidetector CT scans (https://github.com/ferchonavarro/anatomy_guided_contrast_c). This allows minimizing the contrast-induced error in opportunistic bone mineral density measurements.

Use of CT Scans for Abdominal Pain in the ED: Factors in Choice

Overutilization of healthcare resources is a threat to long-term healthcare sustainability and patient outcomes. CTis a costly but efficient means of assessing abdominal pain; however, 97 per cent of ED physicians acknowledge its overutilization. This study sought to understand factors that influence ED providers’ decision regarding CT use in the evaluation of abdominal pain. After evaluating a patient for acute abdominal pain, ED providers filled in a form in which the primary diagnosis and index of suspicion were recorded. Bivariate and multivariate analyses were used to identify predictors of outcomes. The CT scan utilization rate was 54.82 per cent. Whereas 34.11 per cent of CT scans were normal, 30 per cent yielded an acute abdominal pathology. Tenderness and rebound tenderness were positive predictors of high index of suspicion [odds ratio (OR) 2.09 and 2.54, respectively]. These variables were also predictive of obtaining a CT scan [OR 2.64 and 3.41, respectively]. Compared with whites, the index of suspicion was 26 per cent and 56 per cent less likely to be high when patients were black [OR 0.73] or Hispanic [OR 0.44] respectively. Blacks and Hispanics were less likely to have CT scans performed than whites [OR 0.58 and 0.48, respectively]. Leukocytosis significantly affected the index of suspicion for acute abdominal pathology, obtaining a CTscan and the acuity of CTscan diagnosis on multivariate analysis. Patients aged ≥60 years had 2.03 odds of acute CT finding compared with those aged <60 years. There is a need for committed efforts to optimize CT scan utilization and eliminate socioeconomic disparities in health care.

Is the risk of contrast-induced nephropathy a real contraindication to perform intravenous contrast enhanced Computed Tomography for non-traumatic acute abdomen in Emergency Surgery Department?

BACKGROUND

Contrast enhanced Computed Tomography (CCT) is the most used imaging test to investigate acute abdominal clinical conditions, because of its high sensitivity and specificity. It is mandatory to make a correct and prompt diagnosis when life threatening abdominal diseases as mesenteric ischemia are suspected. Contrast medium administration was linked to acute renal failure, therefore radiologist often prefer to perform CCT without contrast in patients needing to undergo the exam with increased serum creatinine.  The aim of the review was to focus on the incidence of contrast induced nephropathy in patients presenting non-traumatic acute abdominal clinical conditions, who underwent CCT with intravenous contrast agent administration in emergency setting.

MATERIALS AND METHODS

The systematic review protocol was guided by the Preferred Reporting Items for Systematic Reviews and Meta-analyses Protocol (PRISMA-P). Quality of the evidence will be evaluated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology.

RESULTS

The strongest currently available evidence on the incidence of post-contrast acute kidney injury (AKI) following intravenous contrast agent administration consists in a meta-analysis of observational studies. Data extracted from meta-analyses demonstrate that, compared with non-contrast CT, CCT was not significantly associated with AKI. Moreover, the risk of AKI (RR=0.79; 95% confidence interval [CI]: 0.62, 1.02; P=.07), death (RR=0.95; 95% CI: 0.55, 1.67; P=.87), and dialysis (RR=0.88; 95% CI: 0.23, 3.43; P=.85) is similar, compared with the risk of AKI in the non-contrast medium group. Furthermore, intravenous low-osmolality iodinated contrast material is a nephrotoxic risk factor, but not in patients with a stable SCr level less than 1.5 mg/dL, therefore many factors other than contrast material could affect PC-AKI rates.

DISCUSSION AND CONCLUSIONS

The benefits of diagnostic information gained from contrast enhanced TC in assessing AA are fundamental in some clinical scenarios. The risk of contrast induced nephropathy (CIN) is negligible in patients with normal renal function but the incidence appears to rise to as high as 25% in patients with pre-existing renal impairment or in the presence of risk factors such as diabetes, advanced age, vascular disease and use of certain concurrent medications. The incidence of CIN/AKI after intravenous contrast administration is very low in general population. Radiologists and referring physicians should be familiar with the risk factors for renal disease, CIN and preventing measures.