To assess if quantitative integrated deep learning and radiomics features can predict the PD-L1 expression level in preoperative MRI of hepatocellular carcinoma (HCC) patients. The data in this study consist of 103 hepatocellular carcinoma patients who received immunotherapy in a single center. These patients were divided into a high PD-L1 expression group (30 patients) and a low PD-L1 expression group (73 patients). Both radiomics and deep learning features were extracted from their MRI sequence of T2-WI, which were merged into an integrative feature space for machine learning for the prediction of PD-L1 expression. The five-fold cross-validation was adopted to validate the performance of the model, while the AUC was used to assess the predictive ability of the model. Based on the five-fold cross-validation, the integrated model achieved the best prediction performance, with an AUC score of 0.897 ± 0.084, followed by the deep learning-based model with an AUC of 0.852 ± 0.043 then the radiomics-based model with AUC of 0.794 ± 0.035. The feature set integrating radiomics and deep learning features is more effective in predicting PD-L1 expression level than only one feature type. The integrated model can achieve fast and accurate prediction of PD-L1 expression status in preoperative MRI of HCC patients.
22 February 2022 in Journal of Medical and Biological Engineering
Embolization is the prevailing therapy for tumor-targeting, anti-organ hyperfunction, and hemostasis. However, the injection of embolic agents largely depends on the experiences of doctors as assisted by X-ray, which will negate the health of the doctor. To avail embolization therapy feasible even in hospitals without experienced doctors and to prevent the doctors from exposion to X-ray, robotization is a promising alternative. To these ends, building the relationship between physiological parameters and hemodynamic parameters during embolization is crucial. This study takes the renal artery-kidney system of rabbits as the model case to investigate the dynamics of vascular embolization by numerical simulation using porous media for injection of embolic agents. The capillaries at the embolic site inside the kidney are modeled as porous media. The flow from the artery to the vein through the porous media is assumed as a viscous resistance fluid. The resistance, which increases with the increasing degree of embolization, is approached by CFD simulations. According to simulation results, a prediction model of flow resistance is established, enabling building the control law of an embolic agents injection robot. Experimental tests provide physical geometries and relevant parameters for the simulations as well as caliber to verify the simulation results. It is demonstrated that the currently proposed prediction model reflects the relationship between embolic agent injection and hemodynamic parameters reliably, enabling quantitative assessment of the degree of embolization with local blood pressure in the artery of the organ.
Objectives: To compare the therapeutic efficacy and safety of balloon-assisted transarterial chemoembolization (B-TACE) versus conventional TACE (C-TACE) in hepatocellular carcinoma (HCC) and to evaluate the potential predictive value of intraoperative balloon-occluded arterial stump pressure (Boasp). Methods: In this prospective, single-centre, randomized controlled study, 60 patients with hepatocellular carcinoma were allocated to either the B-TACE group (n = 30) or the C-TACE group (n = 30). One patient in the B-TACE group was lost to follow-up after allocation. The primary analyses were conducted according to the intention-to-treat (ITT) principle, including all randomized patients, with conservative handling of missing data. Sensitivity analyses were performed to assess the robustness of the results. Tumor response and survival outcomes were evaluated using the modified Response Evaluation Criteria in Solid Tumors (mRECIST) and Cox proportional hazards regression models. Intraoperative balloon-occluded arterial stump pressure (BOASP) was measured as an exploratory parameter to quantify embolization adequacy. Adverse events (AEs) were systematically assessed and graded according to the Common Terminology Criteria for Adverse Events (CTCAE), version 5.0. Results: TACE achieved a higher 3-month ORR (63.3% vs. 10.0%, p < 0.001) and 6-month disease control rates (80.0% vs. 36.7%, p < 0.001), with PFS (HR = 0.30, 95% CI 0.148–0.608) and procedures within 6 months (1 vs. 3, p < 0.001). The 6-month surgical conversion rate was higher (34.5% vs. 6.7%, p = 0.009). Changes in Boasp correlated with efficacy (AUC = 0.825, p = 0.0398). Severe infections were lower in B-TACE (17.2% vs. 76.7%, p < 0.001). Conclusions: B-TACE offers superior efficacy, survival, and surgical conversion versus C-TACE with favorable safety. Boasp provides a quantitative biomarker for predicting treatment response.