RESUMEN
Predicting response to treatment plays a key role to assist radiologists in hepato-cellular carcinoma (HCC) therapy planning. The most widely used treatment for unresectable HCC is the trans-arterial chemoembolization (TACE). A complete radiological response after the first TACE is a reliable predictor of treatment favourable outcome. However, visual inspection of contrast-enhanced CT scans is time-consuming, error prone and too operator-dependent. Thus, in this paper we propose TwinLiverNet: a deep neural network that is able to predict TACE treatment outcome through learning visual cue from CT scans. TwinLiverNet, specifically, integrates 3D convolutions and capsule networks and is designed to process simultaneously late arterial and delayed phases from contrast-enhanced CTs. Experimental results carried out on a dataset consisting of 126 HCC lesions show that TwinLiverNet reaches an average accuracy of 82% in predicting complete response to TACE treatment. Furthermore, combining multiple CT phases (specifically, late arterial and delayed ones) yields a performance increase of over 12 percent points. Finally, the introduction of capsule layers into the model avoids the model to overfit, while enhancing accuracy.Clinical relevance- TwinLiverNet supports radiologists in visual inspection of CT scans to assess TACE treatment outcome, while reducing inter-operator variability.
Asunto(s)
Carcinoma Hepatocelular , Quimioembolización Terapéutica , Neoplasias Hepáticas , Carcinoma Hepatocelular/diagnóstico por imagen , Carcinoma Hepatocelular/terapia , Humanos , Neoplasias Hepáticas/diagnóstico por imagen , Neoplasias Hepáticas/terapia , Tomografía Computarizada por Rayos X , Resultado del TratamientoRESUMEN
OBJECTIVES: To evaluate the incidence and the time of onset of early micro-embolism after CAS (carotid artery stenting) with two different mesh-covered stents and to assess the role of DW-MRI (Diffusion-weighted magnetic resonance imaging) in their prediction. METHODS: Single-institution prospective study including 50 patients (33 male, median age 74 years) who underwent CAS with Roadsaver® or CGuard™. All patients with primary stenosis (37/50, 74%) had carotid plaque DW-MRI pre-procedure, with both qualitative evaluation of the hyperintensity and ADC (apparent diffusion coefficient) measurement of the plaque. All patients had brain DW-MRI pre-procedure, at 1 h, 24 h and 30 days post-procedure to evaluate the appearance of hyperintense lesions over time. Imaging analysis was performed in a double-blinded fashion by two radiologists. RESULTS: There were no statistically significant differences between the two stents both in the incidence at 1 h (P = 0.23) and 24 h (P = 0.36) and in the volume of new DWI hyperintense brain lesions at 24 h (P = 0.27). Thirty-four new asymptomatic lesions in 19 patients (38%) were reported: 4 (11.8%) at 1 h, 30 (88.2%) at 24 h. The 30-day DWI-MR showed complete resolution of all lesions and no evidence of new lesion. The incidence of new lesions at 24 h resulted significantly higher in patients with DWI hyperintense carotid plaques (12/16, 75% vs. 0/21, 0%, P < 0.0001). This result was paralleled by the difference in ADC value (0.83 ± 0.21 vs. 1.42 ± 0.52). CONCLUSION: The majority of early asymptomatic brain lesion occurred during the first 24 h after CAS. Pre-procedure high DWI signal of the plaque was associated with an increased incidence of post-procedure microembolizations.