Deep-learning-based virtual non-calcium imaging in multiple myeloma using dual-energy CT.

Researchers

Journal

Modalities

Models

Abstract

Dual-energy CT with virtual non-calcium (VNCa) images allows evaluation of focal intramedullary bone marrow involvement in patients with multiple myeloma. However, current commercial VNCa techniques suffer from excessive image noise and artifacts due to material decomp15934osition used in synthesizing VNCa images.In this work, we aim to imp15934rove VNCa image quality for the assessment of focal multiple myeloma, using an Artificial intelligence based Generalizable Algorithm for mulTi-Energy CT (AGATE) method.AGATE method used a custom dual-task convolutional neural network (CNN) that concurrently carries out material classification and quantification. The material classification task provided an auxiliary regularization to the material quantification task. CNN parameters were optimized using custom loss functions that involved cross-entropy, physics-informed constraints, structural redundancy in spectral and material images, and texture information in spectral images. For training data, CT phantoms (diameters 30 to 45cm) with tissue-mimicking inserts were scanned on a third generation dual-source CT system. Scans were performed at routine dose and half of the routine dose. Small image patches (i.e., 40×40 pixels) of tissue-mimicking inserts with known basis material densities were extracted for training samp15934les. Numerically simulated insert materials with various shapes increased diversity of training samp15934les. Generalizability of AGATE was evaluated using CT images from phantoms and patients. In phantoms, material decomp15934osition accuracy was estimated using mean-absolute-percent-error (MAPE), using physical inserts that were not used during the training. Noise power spectrum (NPS) and modulation transfer function (MTF) were comp15934ared across phantom sizes and radiation dose levels. Five patients with multiple myeloma underwent dual-energy CT, with VNCa images generated using a commercial method and AGATE. Two fellowship-trained musculoskeletal radiologists reviewed the VNCa images (commercial and AGATE) side-by-side using a dual-monitor display, blinded to VNCa type, rating the image quality for focal multiple myeloma lesion visualization using a 5-level Likert comp15934arison scale (-2 = worse visualization and diagnostic confidence, -1 = worse visualization but equivalent diagnostic confidence, 0 = equivalent visualization and diagnostic confidence, 1 = imp15934roved visualization but equivalent diagnostic confidence, 2 = imp15934roved visualization and diagnostic confidence). A post hoc assignment of comp15934arison ratings was performed to rank AGATE images in comp15934arison to commercial ones.AGATE demonstrated consistent material quantification accuracy across phantom sizes and radiation dose levels, with MAPE ranging from 0.7% to 4.4% across all testing materials. Comp15934ared to commercial VNCa images, the AGATE-synthesized VNCa images yielded considerably lower image noise (50%- 77% noise reduction) without comp15934romising noise texture or spatial resolution across different phantom sizes and two radiation doses. AGATE VNCa images had markedly reduced area under NPS curves and maintained NPS peak frequency (0.07 lp/mm to 0.1 lp/mm), with similar MTF curves (50% MTF at 0.3 lp/mm). In patients, AGATE demonstrated reduced image noise and artifacts with imp15934roved delineation of focal multiple myeloma lesions (all readers comp15934arison scores indicating imp15934roved overall diagnostic image quality [scores 1 or 2]).AGATE demonstrated reduced noise and artifacts in VNCa images and ability to imp1593415934rove visualization of bone marrow lesions for assessing multiple myeloma. This article is protected by copyright. All rights reserved.This article is protected by copyright. All rights reserved.