Artificial intelligence software in pulmonary nodule assessment.

BACKGROUND: This study tests the impact of the addition of autonomous computed tomography (CT) interpreting software to radiologist assessment of pulmonary nodules. METHODS: Computed tomography scans for nodule assessment were identified retrospectively. Lung cancer risk factors, initial radiologist (RAD) report, Philips Lung Nodule software report (computer-aided nodule (CAD)) and radiologist report following the review of CT images and CAD (RAD + CAD) were collected. Follow-up recommendations based on current guidelines were derived from each report. RESULTS: In all, 100 patients were studied. Median maximal diameter of the largest nodule reported by RAD and RAD + CAD were similar at 10.0 and 9.0 mm, respectively (p = 0.06) but were reported as larger by CAD at 11.8 mm (p < 0.001). Follow-up recommendations derived from RAD + CAD were less intensive in 23 (23%) and more intensive in 34 (34%) than that of RAD. DISCUSSION: This study suggests that autonomous software use can alter radiologist assessment of pulmonary nodules such that suggested follow-up is altered.

Computed tomographic pulmonary angiography and pulmonary embolism: predictive value of a d-dimer assay.

BACKGROUND: Computed tomographic pulmonary angiography (CTPA) is increasingly being used as first investigation for suspected pulmonary embolism (PE). The investigation has high predictive value, but is resource and time intensive and exposes patients to considerable radiation. Our aim was to assess the potential value of a negative d-dimer assay to exclude pulmonary emboli and reduce the number of performed CTPAs. METHODS: All CTPAs performed in a Scottish secondary care hospital for a fourteen month period were retrospectively reviewed. Collected data included the presence or absence of PE, d-dimer results and patient demographics. PE positive CTPAs were reviewed by a specialist panel. RESULTS: Pulmonary embolisms were reported for 66/405 (16.3%) CTPAs and d-dimer tests were performed for 216 (53%). 186/216 (86%) patients had a positive and 30 (14%) a negative d-dimer result. The panel agreed 5/66 (7.6%) false positive examinations. The d-dimer assay's negative predictive value was 93.3% (95% CI = 76.5%-98.8%) based on the original number of positive CTPAs and 100% (95% CI = 85.9%-100%) based on expert review. Significant non-PE intrapulmonary pathology was reported for 312/405 (77.0) CTPAs, including 13 new diagnoses of carcinoma. CONCLUSIONS: We found that a low d-dimer score excluded all pulmonary embolisms, after a further specialist panel review identified initial false positive reports. However, current evidence-based guidelines still recommend that clinicians combine a d-dimer result with a validated clinical risk score when selecting suitable patients for CTPA. This may result in better use of limited resources, prevent patients being exposed to unnecessary irradiation and prevent potential complications as a result of iodinated contrast.