AI-Based 3D-Segmentation Quantifies Sarcopenia in Multiple Myeloma Patients

Abstract

Background: Sarcopenia is characterized by a loss of muscle mass and strength, resulting in functional limitations and an increased risk of falls, injuries and fractures. The aim of this study was to obtain detailed information on skeletal muscle changes in patients with multiple myeloma (MM) during treatment. Methods: A total of 51 patients diagnosed with MM who had undergone whole-body low-dose computed tomography acquisition prior to induction therapy (T1) and post autologous stem cell transplantation (T2) were examined retrospectively. Total volume (TV), muscle volume (MV) and intramuscular adipose tissue volume (IMAT) of the autochthonous back muscles, the iliopsoas muscle and the gluteal muscles were evaluated on the basis of the resulting masks of the BOA tool with the fully automated combination of TotalSegmentator and a body composition analysis. An in-house trained artificial intelligence network was used to obtain a fully automated three-dimensional segmentation assessment. Results: Patients' median age was 58 years (IQR 52-66), 38 were male and follow-up CT-scans were performed after a mean of 11.8 months (SD ± 3). Changes in MV and IMAT correlated significantly with Body-Mass-Index (BMI) (r = 0.7, p < 0.0001). Patients (n = 28) with a decrease in BMI (mean -2.2 kg/m²) during therapy lost MV (T1: 3419 cm³, IQR 3176-4000 cm³ vs. T2: 3226 cm³, IQR 3014-3662 cm³, p < 0.0001) whereas patients (n = 20) with an increased BMI (mean +1.4 kg/m²) showed an increase in IMAT (T1: 122 cm³, IQR 96.8-202.8 cm³ vs. T2: 145.5 cm³, IQR 115-248 cm³, p = 0.0002). Loss of MV varied between different muscle groups and was most prominent in the iliopsoas muscle (-9.8%) > gluteus maximus (-9.1%) > gluteus medius (-5.8%) > autochthonous back muscles (-4.3%) > gluteus minimus (-1.5%). Increase in IMAT in patients who gained weight was similar between muscle groups. Conclusions: The artificial intelligence-based three-dimensional segmentation process is a reliable and time-saving method to acquire in-depth information on sarcopenia in MM patients. Loss of MV and increase in IMAT were reliably detectable and associated with changes in BMI. Loss of MV was highest in muscles with more type 2 muscle fibers (fast-twitch, high energy) whereas muscles with predominantly type 1 fibers (slow-twitch, postural control) were less affected. This study provides valuable insight into muscle changes of MM patients during treatment, which might aid in tailoring exercise interventions more precisely to patients' needs.

Keywords: X-ray computed; induction chemotherapy; intelligent systems; multiple myeloma; sarcopenia; tomography.

Figure 1

Example of muscle segmentation. (left) Axial view of whole-body low-dose CT without contrast agent at the level of the first lumbar vertebra. (right) AI-based segmentation of autochthone muscle and psoas muscle (red) with a total volume of 1149 cm³: vertebra in dark grey and bowels in light grey.

Figure 2

Changes in total volume correlated significantly with changes in Body-Mass-Index (BMI), as calculated by Spearman correlation (r = 0.7, p < 0.0001).

Figure 3

Loss of muscle volume in patients with a decrease in BMI. There was a significant loss of volume detectable in all muscles, except for the M. gluteus minimus. Significant differences of loss of volume between different muscles was calculated by Mann–Whitney test (highlighted by asterisk) and significance level was determined as p < 0.05.

Figure 4

Changes of IMAT in patients with an increase in BMI. There was a significant increase detectable in all muscles between T1 and T2. Increase in IMAT was similar in all muscles. (Mann–Whitney test).