Radiomics and Artificial Intelligence for Outcome Prediction in Multiple Myeloma Patients Undergoing Autologous Transplantation: A Feasibility Study with CT Data.

Daniela Schenone,Sara Aquino,Alida Dominietto,Federica Rossi,Lorenzo Torri,Francesco Frassoni,Emanuele Angelucci,Michele Cea,Bianca Bignotti,Alberto Stefano Tagliafico,Cristina Campi,Michele Piana

doi:10.3390/diagnostics11101759

Daniela Schenone, Sara Aquino + Show 10 more

Open Access

https://doi.org/10.3390/diagnostics11101759

Copy DOI

Abstract

Multiple myeloma is a plasma cell dyscrasia characterized by focal and non-focal bone lesions. Radiomic techniques extract morphological information from computerized tomography images and exploit them for stratification and risk prediction purposes. However, few papers so far have applied radiomics to multiple myeloma. A retrospective study approved by the institutional review board: n = 51 transplanted patients and n = 33 (64%) with focal lesion analyzed via an open-source toolbox that extracted 109 radiomics features. We also applied a dedicated tool for computing 24 features describing the whole skeleton asset. The redundancy reduction was realized via correlation and principal component analysis. Fuzzy clustering (FC) and Hough transform filtering (HTF) allowed for patient stratification, with effectiveness assessed by four skill scores. The highest sensitivity and critical success index (CSI) were obtained representing each patient, with 17 focal features selected via correlation with the 24 features describing the overall skeletal asset. These scores were higher than the ones associated with a standard cytogenetic classification. The Mann–Whitney U-test showed that three among the 17 imaging descriptors passed the null hypothesis. This AI-based interpretation of radiomics features stratified relapsed and non-relapsed MM patients, showing some potentiality for the determination of the prognostic image-based biomarkers in disease follow-up.

Highlights

Plasma cell dyscrasias (PCDs) include monoclonal gammopathy of undetermined significance (MGUS), smoldering multiple myeloma (SMM), and full-blown multiple myeloma (MM) [1]
The presence of either single or multiple bone lesions is a typical signature of MM, which is related to the proliferation of tumor cells from a single clone, so that the unbalanced activation of osteoclasts erodes the medullary and even the cortical bone [6].the CRAB criteria of International Myeloma Working Group (IMWG) underlines the importance of imaging for MM assessment, and recent staging systems rely on the use of imaging modalities like magnetic resonance imaging (MRI), computerized tomography (CT) and hybrid positron emission tomography with CT (PET/CT) [3,4,6,7,8,9,10,11,12,13,14,15]
Focal lesion searching led to the selection of 33/51 (65%) patients whose imaging data were considered for our computational analysis

Summary

Introduction

Plasma cell dyscrasias (PCDs) include monoclonal gammopathy of undetermined significance (MGUS), smoldering multiple myeloma (SMM), and full-blown multiple myeloma (MM) [1]. Around 10% of the SMM population evolves into full-blown MM, whose early mortality is nowadays around 28% five years after diagnosis [2]. The presence of either single or multiple bone lesions is a typical signature of MM, which is related to the proliferation of tumor cells from a single clone, so that the unbalanced activation of osteoclasts erodes the medullary and even the cortical bone [6].the CRAB criteria of IMWG underlines the importance of imaging for MM assessment, and recent staging systems rely on the use of imaging modalities like magnetic resonance imaging (MRI), computerized tomography (CT) and hybrid positron emission tomography with CT (PET/CT) [3,4,6,7,8,9,10,11,12,13,14,15]. Just the availability of different imaging modalities and the high variability of image interpretation imply a notable heterogeneity as far as the use of imaging for MM clinical practice is concerned [6,12,16]

Objectives

Methods

Results

Conclusion