A multitask deep learning radiomics model for predicting the macrotrabecular-massive subtype and prognosis of hepatocellular carcinoma after hepatic arterial infusion chemotherapy

Abstract

Background The macrotrabecular-massive (MTM) is a special subtype of hepatocellular carcinoma (HCC), which has commonly a dismal prognosis. This study aimed to develop a multitask deep learning radiomics (MDLR) model for predicting MTM and HCC patients’ prognosis after hepatic arterial infusion chemotherapy (HAIC). Methods From June 2018 to March 2020, 158 eligible patients with HCC who underwent surgery were retrospectively enrolled in MTM related cohorts, and 752 HCC patients who underwent HAIC were included in HAIC related cohorts during the same period. DLR features were extracted from dual-phase (arterial phase and venous phase) contrast-enhanced computed tomography (CECT) of the entire liver region. Then, an MDLR model was used for the simultaneous prediction of the MTM subtype and patient prognosis after HAIC. The MDLR model for prognostic risk stratification incorporated DLR signatures, clinical variables and MTM subtype. Findings The predictive performance of the DLR model for the MTM subtype was 0.968 in the training cohort [TC], 0.912 in the internal test cohort [ITC] and 0.773 in the external test cohort [ETC], respectively. Multivariable analysis identified portal vein tumor thrombus (PVTT) (p = 0.012), HAIC response (p < 0.001), HAIC sessions (p < 0.001) and MTM subtype (p < 0.001) as indicators of poor prognosis. After incorporating DLR signatures, the MDLR model yielded the best performance among all models (AUC, 0.855 in the TC, 0.805 in the ITC and 0.792 in the ETC). With these variables, the MDLR model provided two risk strata for overall survival (OS) in the TC: low risk (5-year OS, 44.9%) and high risk (5-year OS, 4.9%). Interpretation A tool based on MDLR was developed to consider that the MTM is an important prognosis factor for HCC patients. MDLR showed outstanding performance for the prognostic risk stratification of HCC patients who underwent HAIC and may help physicians with therapeutic decision making and surveillance strategy selection in clinical practice.

An Informative Review of Radiomics Studies on Cancer Imaging: The Main Findings, Challenges and Limitations of the Methodologies

“…This is critical not only by a medico-legal point of view, but also by a "human" point of view, because, like Coppola et al [81] stressed, we must not superintend the meaning of the irreplaceable doctor-patient bond. Connecting doctors and patients directly will always be an important phase of healthcare services that artificial intelligence can never replace [2,20,68,[102][103][104][105][106][107][108][109].…”

“…Radiomics is the analysis of medical images to obtain multiple quantitative data that cannot be identified by the human eye [1][2][3][4][5][6][7][8]. It provides insight into underlying pathophysiological phenomena not accessible to simple visual analysis.…”

“…In medicine, handcrafted radiomic models use data analytics to extract many features from medical images and is made up of several steps: (1) segmentation of the target lesion with manual segmentation by radiologists or with automatic and semi-automatic tools, (2) feature extraction to obtain multiple quantitative metrics and parameters from medical images, (3) feature selection with the aim of reducing the number of extracted features by avoiding correlated or redundant metrics, (4) analysis/classification by creating a predictive model using machine and deep learning approaches and (5) the validation of the results [9][10][11][12][13][14][15][16][17][18].…”

Machine Learning and Radiomics Analysis for Tumor Budding Prediction in Colorectal Liver Metastases Magnetic Resonance Imaging Assessment

“…So, patients with liver metastases with tumor budding and/or KRAS mutational status [39,40] respond poorly to anti-EGFR therapy [32]. In the context of personalized medicine, this is evident as the possibility to predict several prognostic markers allows us to identify the best treatment for a specific patient [41][42][43][44][45][46]. Radiomics analysis could be a promising tool to evaluate a lesion "virtually", with the possibility to analyze the whole tumor during the disease history to obtain those markers which can affect the treatment choice [47][48][49][50][51][52][53][54][55][56][57][58][59][60][61][62][63][64][65].…”

Scientific Status Quo of Small Renal Lesions: Diagnostic Assessment and Radiomics

“…In addition, it should be considered that up to 20% of SRMs are benign [15] and that the risk of malignancy rises with increasing size [20][21][22]. For these reasons, an adequate characterization of renal lesions through the different imaging techniques is critical to optimize SRM management and to improve patient outcomes, preserving renal function in the best viable way and avoiding the risk of overtreatment [23][24][25][26][27][28][29][30].…”