Smart chest X-ray worklist prioritization using artificial intelligence: a clinical workflow simulation.

OBJECTIVE: The aim is to evaluate whether smart worklist prioritization by artificial intelligence (AI) can optimize the radiology workflow and reduce report turnaround times (RTATs) for critical findings in chest radiographs (CXRs). Furthermore, we investigate a method to counteract the effect of false negative predictions by AI-resulting in an extremely and dangerously long RTAT, as CXRs are sorted to the end of the worklist. METHODS: We developed a simulation framework that models the current workflow at a university hospital by incorporating hospital-specific CXR generation rates and reporting rates and pathology distribution. Using this, we simulated the standard worklist processing "first-in, first-out" (FIFO) and compared it with a worklist prioritization based on urgency. Examination prioritization was performed by the AI, classifying eight different pathological findings ranked in descending order of urgency: pneumothorax, pleural effusion, infiltrate, congestion, atelectasis, cardiomegaly, mass, and foreign object. Furthermore, we introduced an upper limit for the maximum waiting time, after which the highest urgency is assigned to the examination. RESULTS: The average RTAT for all critical findings was significantly reduced in all prioritization simulations compared to the FIFO simulation (e.g., pneumothorax: 35.6 min vs. 80.1 min; p < 0.0001), while the maximum RTAT for most findings increased at the same time (e.g., pneumothorax: 1293 min vs 890 min; p < 0.0001). Our "upper limit" substantially reduced the maximum RTAT in all classes (e.g., pneumothorax: 979 min vs. 1293 min/1178 min; p < 0.0001). CONCLUSION: Our simulations demonstrate that smart worklist prioritization by AI can reduce the average RTAT for critical findings in CXRs while maintaining a small maximum RTAT as FIFO. KEY POINTS: • Development of a realistic clinical workflow simulator based on empirical data from a hospital allowed precise assessment of smart worklist prioritization using artificial intelligence. • Employing a smart worklist prioritization without a threshold for maximum waiting time runs the risk of false negative predictions of the artificial intelligence greatly increasing the report turnaround time. • Use of a state-of-the-art convolution neural network can reduce the average report turnaround time almost to the upper limit of a perfect classification algorithm (e.g., pneumothorax: 35.6 min vs. 30.4 min).

Hepatic Artery Delineation on Ultrasound Volumes Comparing B-Flow and Color Doppler for Postoperative Monitoring of Pediatric Liver Transplants.

(1) Background: Accurate hepatic artery (HA) depiction following pediatric liver transplantation (LT) is essential for graft surveillance but challenging on ultrasound (US). This study assesses if improved HA delineation can be achieved by recording two-dimensional US volumes in Color Doppler (CD) and B-flow technique. (2) Methods: Of 42 consecutive LT, 37 cases were included, and HA delineation was retrospectively rated using a four-point score (0 = HA not detectable, 3 = HA fully detectable, separable from portal vein) within 48 h post-LT (U1) and before discharge (U2). (3) Results: Adding B-flow compared with CD alone showed superior results at neohilar (U1: 2.2 ± 1.0 vs. 1.1 ± 0.8, p < 0.0001; U2: 2.5 ± 0.8 vs. 1.5 ± 0.9, p < 0.0001) and segmental levels (U1: 2.8 ± 0.6 vs. 0.6 ± 0.8, p < 0.0001; U2: 2.8 ± 0.6 vs. 0.7 ± 0.5, p < 0.0001). (4) Conclusions: Standardized US volume recordings combining B-flow and CD can effectively delineate the HA along its vascular course in pediatric LT. The technique should be further evaluated as a standard monitoring instrument to rule out vascular complications after LT.

Receptor tyrosine kinase MET as potential target of multi-kinase inhibitor and radiosensitizer sorafenib in HNSCC.

BACKGROUND: The multi-kinase inhibitor sorafenib displays antitumoral effects in head and neck squamous cell carcinoma (HNSCC); however, the targeted kinases are unknown. Here we aimed to identify those kinases to determine the mechanism of sorafenib-mediated effects and establish candidate biomarkers for patient stratification. METHODS: The effects of sorafenib and MET inhibitors crizotinib and SU11274 were analyzed using a slide-based antibody array, Western blotting, proliferation, and survival assays. X-rays were used for irradiations. RESULTS: Sorafenib inhibited auto-phosphorylation of epidermal growth factor receptor and MET, which has not been described previously. MET expression in HNSCC cells was not always associated with activity/phosphorylation. Furthermore, sorafenib-dependent cell kill and radiosensitization was not associated with MET level. Although MET inhibitors blocked proliferation, they caused only mild cytotoxicity and no radiosensitization. CONCLUSION: We identified MET as a new potential target of sorafenib. However, MET inhibition is not the cause for sorafenib-mediated cytotoxicity or radiosensitization.

Effect of sorafenib on cisplatin-based chemoradiation in head and neck cancer cells.

Despite aggressive chemoradiation (CRT) protocols in the treatment of patients with head and neck squamous cell carcinomas (HNSCC), the outcome is still unfavorable. To improve therapy efficacy we had already successfully tested the multikinase inhibitor sorafenib in combination with irradiation (IR) in previous studies on HNSCC cell lines. In this study we investigated its effect on combined CRT treatment using cisplatin.Radio- and chemosensitivity with and without sorafenib was measured in four HNSCC cell lines and normal fibroblasts (NF) by colony formation assay. Apoptosis and cell cycle analysis were performed by flow cytometry.In HNSCC cells, sorafenib enhanced the antiproliferative effect of cisplatin without affecting apoptosis induction and with only minor effects on cell inactivation. Sorafenib added prior to irradiation enhanced cellular radiosensitivity in three of the tested HNSCC cell lines and caused massive overall cell inactivation when combined with CRT. In contrast, sorafenib did not radiosensitize NF and reduced cisplatin-induced cell inactivation. Cell inactivation by IR and cisplatin is further increased by the addition of sorafenib in HNSCC, but not in NF cells. Therefore, sorafenib is a promising candidate to improve therapy efficacy for HNSCC.

Sorafenib sensitizes head and neck squamous cell carcinoma cells to ionizing radiation.

BACKGROUND AND PURPOSE: There is a great need to improve the outcome of locoregionally advanced squamous cell carcinomas of the head and neck (HNSCC). Standard treatment includes a combination of surgery, radio- and chemotherapy. The addition of molecular targeting agents to conventional treatment may improve outcomes. In this study the Raf inhibitor sorafenib was used to increase the radiosensitivity of HNSCC cell lines. MATERIAL AND METHODS: In a panel of six cell lines (A549, FaDu, UTSCC 60A, UTSCC 42A, UTSCC 42B, UTSCC 29) radiosensitivity was measured by colony formation assay and apoptosis and cell cycle analysis were performed by flow cytometry. DNA repair was analyzed by 53BP1 immunohistochemistry. RESULTS: Sorafenib added prior to irradiation resulted in an increased cellular radiosensitivity (DEF0.5=1.11-1.84). Radiosensitization was not caused by an enhanced rate of apoptosis or cell cycle effects. In contrast, sorafenib was shown for the first time to block the repair of DNA double-strand breaks (DSB). CONCLUSION: Our data suggest that sorafenib may be used to overcome the radioresistance of HNSCC through the inhibition of DSB repair.