Computed tomography complements ultrasound for the differential diagnosis of traumatic neuroma from recurrent tumor in patients with postoperative thyroid cancer

CT-based quantification of trachea shape to detect invasion by thyroid cancer

OBJECTIVE

This study aims to develop a CT-based method for quantifying tracheal shape and evaluating its ability to distinguish between cases with or without tracheal invasion in patients with thyroid carcinoma.

METHODS

A total of 116 quantitative shape features, including 56 geometric moments and 60 bounding shape features, were defined. The tracheal lumen was semi-automatically defined with a CT threshold of less than - 500 HU. Three contiguous slices with the 1st, 2nd, and 3rd smallest trachea lumen areas were contiguously selected, and the appropriate number of slices to be included was determined. Fifty-six patients with differentiated thyroid carcinoma (DTC) invading the trachea and 22 patients with DTC but without invasion were retrospectively included. A receiver operating characteristic (ROC) curve was applied to select the representative shape features and determine the optimal threshold.

RESULTS

23.3%, 25.9%, and 24.1% of the features displayed an area under the ROC curve (AUC) ≥ 0.800 when derived from 1, 2, and 3 slices, respectively. Calculating feature values from two slices with the 1st and 2nd smallest tracheal lumen area were considered appropriate. Six final features, including 3 geometric moments and 3 bounding shape features, were selected to determine the tracheal invasion status of DTC and displayed AUCs of 0.875-0.918, accuracies of 0.821-0.891, sensitivities of 0.813-0.893, and specificities of 0.818-0.932, outperforming the visual evaluation results.

CONCLUSIONS

Geometric moments and bounding shape features can quantify the tracheal shape and are reliable for identifying DTC tracheal invasion. The selected features quantified the extent of tracheal deformity in DTC patients with and without tracheal invasion.

CLINICAL RELEVANCE STATEMENT

Six geometric features provide a non-invasive, semi-automated evaluation of the tracheal invasion status of thyroid cancer.

KEY POINTS

• A novel method for quantifying tracheal shape using 56 geometric moments and 60 bounding shape features was developed. • Six features identify tracheal invasion by thyroid carcinoma. • The selected features quantified the extent of tracheal deformity in differentiated thyroid carcinoma patients with and without tracheal invasion.

Efficacy of percutaneous image-guided biopsy for diagnosis of intrahepatic cholangiocarcinoma

PURPOSE

To evaluate the efficacy of percutaneous biopsy for diagnosing intrahepatic cholangiocarcinoma (IHCCA).

METHODS

Retrospective review of biopsy and pathology databases from 2006 to 2019 yielded 112 patients (54F/58 M; mean age, 62.9 years; 27 cirrhotic) with IHCCA who underwent percutaneous biopsy. Data regarding the lesion, biopsy procedure technique, and diagnostic yield were collected. If biopsy was non-diagnostic or discordant with imaging, details of repeat biopsy or resection/explant were gathered. A control group of 100 consecutive patients (56F/44 M; mean age, 63 years, 5 cirrhotic) with focal liver lesions > 1 cm was similarly assessed.

RESULTS

Mean IHCCA lesion size was 6.1 ± 3.6 cm, with dominant lesion sampled in 78% (vs. satellite in 22%). 95% (n = 106) were US guided and 96% were core biopsies (n = 108), typically 18G (n = 102, 91%), median 2 passes. 18 patients (16%) had discordant/ambiguous pathology results requiring repeat biopsy, with two patients requiring 3-4 total attempts. A 4.4% minor complication rate was seen. Mean time from initial biopsy to final diagnosis was 60 ± 120 days. Control group had mean lesion size of 2.9 ± 2.5 cm and showed a non-diagnostic rate of 3.3%, both significantly lower than that seen with CCA, with average time to diagnosis of 21 ± 28.8 days (p = 0.002, p = 0.001).

CONCLUSION

IHCCA is associated with lower diagnostic yield at initial percutaneous biopsy, despite larger target lesion size. If a suspicious lesion yields a biopsy result discordant with imaging, the radiologist should recommend prompt repeat biopsy to prevent delay in diagnosis.

Diagnostic performance of artificial intelligence approved for adults for the interpretation of pediatric chest radiographs

Artificial intelligence (AI) applied to pediatric chest radiographs are yet scarce. This study evaluated whether AI-based software developed for adult chest radiographs can be used for pediatric chest radiographs. Pediatric patients (≤ 18 years old) who underwent chest radiographs from March to May 2021 were included retrospectively. An AI-based lesion detection software assessed the presence of nodules, consolidation, fibrosis, atelectasis, cardiomegaly, pleural effusion, pneumothorax, and pneumoperitoneum. Using the pediatric radiologist’s results as standard reference, we assessed the diagnostic performance of the software. For the total 2273 chest radiographs, the AI-based software showed a sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of 67.2%, 91.1%, 57.7%, 93.9%, and 87.5%, respectively. Age was a significant factor for incorrect results (odds radio 0.821, 95% confidence interval 0.791–0.851). When we excluded cardiomegaly and children 2 years old or younger, sensitivity, specificity, PPV, NPV and accuracy significantly increased (86.4%, 97.9%, 79.7%, 98.7% and 96.9%, respectively, all p < 0.001). In conclusion, AI-based software developed with adult chest radiographs showed diagnostic accuracies up to 96.9% for pediatric chest radiographs when we excluded cardiomegaly and children 2 years old or younger. AI-based lesion detection software needs to be validated in younger children.

Diagnostic effect of artificial intelligence solution for referable thoracic abnormalities on chest radiography: a multicenter respiratory outpatient diagnostic cohort study

Objectives

We aim ed to evaluate a commercial artificial intelligence (AI) solution on a multicenter cohort of chest radiographs and to compare physicians' ability to detect and localize referable thoracic abnormalities with and without AI assistance.

Methods

In this retrospective diagnostic cohort study, we investigated 6,006 consecutive patients who underwent both chest radiography and CT. We evaluated a commercially available AI solution intended to facilitate the detection of three chest abnormalities (nodule/masses, consolidation, and pneumothorax) against a reference standard to measure its diagnostic performance. Moreover, twelve physicians, including thoracic radiologists, board-certified radiologists, radiology residents, and pulmonologists, assessed a dataset of 230 randomly sampled chest radiographic images. The images were reviewed twice per physician, with and without AI, with a 4-week washout period. We measured the impact of AI assistance on observer's AUC, sensitivity, specificity, and the area under the alternative free-response ROC (AUAFROC).

Results

In the entire set (n = 6,006), the AI solution showed average sensitivity, specificity, and AUC of 0.885, 0.723, and 0.867, respectively. In the test dataset (n = 230), the average AUC and AUAFROC across observers significantly increased with AI assistance (from 0.861 to 0.886; p = 0.003 and from 0.797 to 0.822; p = 0.003, respectively).

Conclusions

The diagnostic performance of the AI solution was found to be acceptable for the images from respiratory outpatient clinics. The diagnostic performance of physicians marginally improved with the use of AI solutions. Further evaluation of AI assistance for chest radiographs using a prospective design is required to prove the efficacy of AI assistance.

Key Points

• AI assistance for chest radiographs marginally improved physicians’ performance in detecting and localizing referable thoracic abnormalities on chest radiographs.

• The detection or localization of referable thoracic abnormalities by pulmonologists and radiology residents improved with the use of AI assistance.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-021-08397-5.