Maria Elena Susi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Maria Elena Susi has a total of 56 co-authors. Here are the top co-authors by number of shared publications:
- Federico Pistoia
IRCCS Ospedale Policlinico San Martino Genoa Italy
5 shared publications - Marta Macciò
Department of Health Sciences (DISSAL), Radiology Section, University of Genova, Genoa, Italy
5 shared publications - Riccardo Picasso
IRCCS Ospedale Policlinico San Martino Genoa Italy
5 shared publications - Carlo Martinoli
University of Genova
4 shared publications - Federico Zaottini
IRCCS Azienda Ospedaliera Metropolitana, 16132 Genoa, Italy
4 shared publications - Giovanni Marcenaro
Department of Health Sciences (DISSAL) University of Genoa Genoa Italy
4 shared publications
Latest Publications
- Open Access
- 14 October 2026
- Published by PAGEPress Publications in Bleeding, Thrombosis and Vascular Biology
- Vol. 5 (s2) , 1-8
CO-21 | FUSION 2: VALIDATION AND CLINICAL PERFORMANCE OF AN INTEGRATED ULTRASOUND–FUNCTIONAL ASSESSMENT SYSTEM IN HEMOPHILIA
Background and Aims. FUSION (Functional & Ultrasound Integrated Observation in Hemophilia) is an integrated musculoskeletal assessment framework combining point-of-care ultrasound (HEAD-US) with a simplified functional evaluation. While ultrasound and clinical joint scores provide complementary information, the extent to which functional performance contributes independent information beyond established tools remains unclear. This study aimed to validate the FUSION composite score and investigate whether its functional component captures clinically relevant dimensions not explained by HEAD-US or Hemophilia Joint Health Score (HJHS) 2.1 alone.Materials and Methods. We conducted a prospective multicenter cross-sectional study including 121 patients with hemophilia A, hemophilia B, or von Willebrand disease undergoing integrated assessment with HEAD-US, HJHS 2.1, and FUSION. The FUSION score combines HEAD-US evaluation of elbows, knees, and ankles with a weighted simplified functional score (F×4) derived from six standardized motor tasks. Convergent validity was assessed using Spearman correlation and Steiger’s z-test. Variance decomposition and residual analyses were performed to identify the proportion and nature of information uniquely captured by FUSION.Results. FUSION showed strong convergent validity with HJHS (ρ=0.764, p<0.001), comparable to the correlation observed between HEAD-US and HJHS (ρ=0.776, p<0.001; Steiger p=0.82). However, the addition of the functional component significantly increased explained variance beyond HEAD-US alone (ΔR²=0.070, p<0.001), supporting an independent contribution of functional assessment. Variance decomposition demonstrated that 25.5% of the variance captured by FUSION was not explained by HJHS. Notably, 43.1% of the variance of the functional component (F×4) remained unexplained by HJHS, indicating that FUSION captures a substantial non-overlapping functional domain. Discordance analyses further supported the complementary role of FUSION. Among patients with HJHS=0, FUSION identified abnormalities in 53.8% of cases. In patients with HEAD-US=0, FUSION detected abnormalities in 55.0% versus 30.0% with HJHS alone (p=0.013), with all additional detections driven by the functional component.Conclusions. FUSION preserves strong alignment with established hemophilia joint assessment tools while simultaneously identifying a substantial and clinically relevant functional domain not captured by HJHS or structural ultrasound evaluation alone. These findings support the concept of a multidimensional structural-functional assessment model and suggest that current joint-focused scores may underestimate global biomechanical and compensatory impairment in hemophilia.
- 14 October 2026
- Published by PAGEPress Publications in Bleeding, Thrombosis and Vascular Biology
- Vol. 5 (s2) , 1-8
Open AccessCO-21 | FUSION 2: VALIDATION AND CLINICAL PERFORMANCE OF AN INTEGRATED ULTRASOUND–FUNCTIONAL ASSESSMENT SYSTEM IN HEMOPHILIA
Background and Aims. FUSION (Functional & Ultrasound Integrated Observation in Hemophilia) is an integrated musculoskeletal assessment framework combining point-of-care ultrasound (HEAD-US) with a simplified functional evaluation. While ultrasound and clinical joint scores provide complementary information, the extent to which functional performance contributes independent information beyond established tools remains unclear. This study aimed to validate the FUSION composite score and investigate whether its functional component captures clinically relevant dimensions not explained by HEAD-US or Hemophilia Joint Health Score (HJHS) 2.1 alone.Materials and Methods. We conducted a prospective multicenter cross-sectional study including 121 patients with hemophilia A, hemophilia B, or von Willebrand disease undergoing integrated assessment with HEAD-US, HJHS 2.1, and FUSION. The FUSION score combines HEAD-US evaluation of elbows, knees, and ankles with a weighted simplified functional score (F×4) derived from six standardized motor tasks. Convergent validity was assessed using Spearman correlation and Steiger’s z-test. Variance decomposition and residual analyses were performed to identify the proportion and nature of information uniquely captured by FUSION.Results. FUSION showed strong convergent validity with HJHS (ρ=0.764, p<0.001), comparable to the correlation observed between HEAD-US and HJHS (ρ=0.776, p<0.001; Steiger p=0.82). However, the addition of the functional component significantly increased explained variance beyond HEAD-US alone (ΔR²=0.070, p<0.001), supporting an independent contribution of functional assessment. Variance decomposition demonstrated that 25.5% of the variance captured by FUSION was not explained by HJHS. Notably, 43.1% of the variance of the functional component (F×4) remained unexplained by HJHS, indicating that FUSION captures a substantial non-overlapping functional domain. Discordance analyses further supported the complementary role of FUSION. Among patients with HJHS=0, FUSION identified abnormalities in 53.8% of cases. In patients with HEAD-US=0, FUSION detected abnormalities in 55.0% versus 30.0% with HJHS alone (p=0.013), with all additional detections driven by the functional component.Conclusions. FUSION preserves strong alignment with established hemophilia joint assessment tools while simultaneously identifying a substantial and clinically relevant functional domain not captured by HJHS or structural ultrasound evaluation alone. These findings support the concept of a multidimensional structural-functional assessment model and suggest that current joint-focused scores may underestimate global biomechanical and compensatory impairment in hemophilia.
- Open Access
- 5 August 2026
- Published by Wiley in Muscle & Nerve
- Vol. 74 (4) , 1256-1262
Imaging of Facial Muscles in Facioscapulohumeral Muscular Dystrophy: An Exploratory Study Comparing Magnetic Resonance Imaging and Ultrasound
Aims: Facioscapulohumeral muscular dystrophy (FSHD) is a genetic progressive muscle disorder often presenting with facial weakness. However, imaging studies specifically evaluating facial muscle involvement and its relationship with clinical severity remain limited. This preliminary study explored magnetic resonance imaging (MRI) and ultrasound (US) assessment of facial muscle involvement and correlations with clinical severity. Methods: This single‐center prospective study enrolled 12 genetically confirmed FSHD1 patients (mean age 57.8 years). Clinical evaluation included the FSHD score, Comprehensive Clinical Evaluation Form (CCEF), and Facial Weakness Score (FWS). Muscle thickness of the orbicularis oris (OOr) and zygomaticus major (ZMj) was measured on both MRI and US, while MRI was additionally used for qualitative assessment of muscle bulk. Inter‐rater reliability was assessed for imaging measurements. Results: MRI and US showed comparable facial muscle thickness, with mean ZMj thickness of 2.34 ± 0.75 mm and 2.12 ± 0.55 mm, and OOr thickness of 2.28 ± 0.70 mm and 2.20 ± 0.67 mm, respectively. Inter‐rater reliability for MRI was high (ICC = 0.84), and reduced ZMj thickness showed moderate to strong correlation with greater disease severity ( ρ up to 0.87, p < 0.001) and impaired facial gestures ( ρ = 0.50–0.77, p < 0.05). Discussion: Despite the small sample size, MRI, and US assessment of facial muscles correlated with clinical severity and functional outcomes in FSHD, supporting their use as complementary tools. Longitudinal studies are required to establish imaging's role in monitoring changes over time.
- 5 August 2026
- Published by Wiley in Muscle & Nerve
- Vol. 74 (4) , 1256-1262
Open AccessImaging of Facial Muscles in Facioscapulohumeral Muscular Dystrophy: An Exploratory Study Comparing Magnetic Resonance Imaging and Ultrasound
Aims: Facioscapulohumeral muscular dystrophy (FSHD) is a genetic progressive muscle disorder often presenting with facial weakness. However, imaging studies specifically evaluating facial muscle involvement and its relationship with clinical severity remain limited. This preliminary study explored magnetic resonance imaging (MRI) and ultrasound (US) assessment of facial muscle involvement and correlations with clinical severity. Methods: This single‐center prospective study enrolled 12 genetically confirmed FSHD1 patients (mean age 57.8 years). Clinical evaluation included the FSHD score, Comprehensive Clinical Evaluation Form (CCEF), and Facial Weakness Score (FWS). Muscle thickness of the orbicularis oris (OOr) and zygomaticus major (ZMj) was measured on both MRI and US, while MRI was additionally used for qualitative assessment of muscle bulk. Inter‐rater reliability was assessed for imaging measurements. Results: MRI and US showed comparable facial muscle thickness, with mean ZMj thickness of 2.34 ± 0.75 mm and 2.12 ± 0.55 mm, and OOr thickness of 2.28 ± 0.70 mm and 2.20 ± 0.67 mm, respectively. Inter‐rater reliability for MRI was high (ICC = 0.84), and reduced ZMj thickness showed moderate to strong correlation with greater disease severity ( ρ up to 0.87, p < 0.001) and impaired facial gestures ( ρ = 0.50–0.77, p < 0.05). Discussion: Despite the small sample size, MRI, and US assessment of facial muscles correlated with clinical severity and functional outcomes in FSHD, supporting their use as complementary tools. Longitudinal studies are required to establish imaging's role in monitoring changes over time.
- Open Access
- 28 May 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (11) , 4168
Efficacy of Bimekizumab on Palmoplantar Psoriasis: A 16-Week Multicenter Retrospective Study—IL PSO (Italian Landscape PSOriasis)
Background/Objectives: Palmoplantar psoriasis (PP) is a challenging variant of psoriasis that affects high-impact areas such as palms and soles and significantly impairs quality of life despite often limited body surface involvement. Conventional topical and systemic therapies may be insufficient, and evidence on biologic treatments for this specific phenotype remains limited. Bimekizumab (BKZ), a monoclonal antibody targeting IL-17A and IL-17F, has shown high efficacy in plaque psoriasis. This study aimed to evaluate the real-world effectiveness and rapidity of action of BKZ in patients with palmoplantar psoriasis compared with patients with psoriasis vulgaris (PV). Methods: We conducted a multicenter retrospective cohort study using data from 22 Italian dermatological units within the IL-PSO (Italian Landscape-Psoriasis) database. Adult patients treated with BKZ between November 2022 and October 2024 were included and categorized into three groups: isolated PP, PP associated with PV (PP + PV), and PV without palmoplantar involvement. Clinical outcomes included the Psoriasis Area and Severity Index (PASI), Dermatology Life Quality Index (DLQI), and pruritus Visual Analogue Scale (VAS). Outcomes were assessed at baseline, week 4, and week 16. Results: A total of 47 patients were included. The baseline PASI was lower in the PP group compared with the PP + PV and PV groups, whereas the DLQI was highest in patients with isolated PP. Rapid clinical improvement was observed in all groups. The mean % PASI reduction at week 4 was 60.5%, 65.1%, and 77.0% in the PP, PP + PV, and PV groups, respectively, increasing to 94.8%, 90.4%, and 91.8% at week 16. The proportion of patients achieving complete clearance (PASI = 0) at week 16 was 73.3% (11/15), 68.2% (15/22), and 70.0% (7/10), respectively. Significant improvements were also observed in DLQI and pruritus scores over time. No significant safety concerns emerged. Conclusions: In this real-world multicenter cohort, bimekizumab demonstrated rapid and high efficacy in patients with palmoplantar psoriasis, both isolated and associated with psoriasis vulgaris. These findings support the use of BKZ as an effective therapeutic option for psoriasis involving high-impact areas, as the palmoplantar, although larger studies are needed to confirm these results.
- 28 May 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (11) , 4168
Open AccessEfficacy of Bimekizumab on Palmoplantar Psoriasis: A 16-Week Multicenter Retrospective Study—IL PSO (Italian Landscape PSOriasis)
Background/Objectives: Palmoplantar psoriasis (PP) is a challenging variant of psoriasis that affects high-impact areas such as palms and soles and significantly impairs quality of life despite often limited body surface involvement. Conventional topical and systemic therapies may be insufficient, and evidence on biologic treatments for this specific phenotype remains limited. Bimekizumab (BKZ), a monoclonal antibody targeting IL-17A and IL-17F, has shown high efficacy in plaque psoriasis. This study aimed to evaluate the real-world effectiveness and rapidity of action of BKZ in patients with palmoplantar psoriasis compared with patients with psoriasis vulgaris (PV). Methods: We conducted a multicenter retrospective cohort study using data from 22 Italian dermatological units within the IL-PSO (Italian Landscape-Psoriasis) database. Adult patients treated with BKZ between November 2022 and October 2024 were included and categorized into three groups: isolated PP, PP associated with PV (PP + PV), and PV without palmoplantar involvement. Clinical outcomes included the Psoriasis Area and Severity Index (PASI), Dermatology Life Quality Index (DLQI), and pruritus Visual Analogue Scale (VAS). Outcomes were assessed at baseline, week 4, and week 16. Results: A total of 47 patients were included. The baseline PASI was lower in the PP group compared with the PP + PV and PV groups, whereas the DLQI was highest in patients with isolated PP. Rapid clinical improvement was observed in all groups. The mean % PASI reduction at week 4 was 60.5%, 65.1%, and 77.0% in the PP, PP + PV, and PV groups, respectively, increasing to 94.8%, 90.4%, and 91.8% at week 16. The proportion of patients achieving complete clearance (PASI = 0) at week 16 was 73.3% (11/15), 68.2% (15/22), and 70.0% (7/10), respectively. Significant improvements were also observed in DLQI and pruritus scores over time. No significant safety concerns emerged. Conclusions: In this real-world multicenter cohort, bimekizumab demonstrated rapid and high efficacy in patients with palmoplantar psoriasis, both isolated and associated with psoriasis vulgaris. These findings support the use of BKZ as an effective therapeutic option for psoriasis involving high-impact areas, as the palmoplantar, although larger studies are needed to confirm these results.
- Open Access
- 19 May 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (10) , 3915
CT of Soft Tissue Infections: Current and Future Perspectives from Diagnosis to Treatment
Background/Objectives: Skin and soft-tissue infections represent a critical diagnostic challenge in the emergency department, requiring rapid differentiation between superficial processes and life-threatening necrotizing conditions. Methods: This clinical update analyzes the central role of CT in the assessment of these pathologies, proposing a management-oriented algorithm to support risk stratification and therapeutic planning. Results: By reviewing the fundamental semiological features of contrast-enhanced CT, this work highlights how the identification of critical markers directly guides the choice between medical management and urgent surgical intervention. In addition to consolidating protocols based on conventional CT, the article examines current perspectives and clinical limitations of emerging technologies, including Spectral CT (DECT and PCD-CT) and artificial intelligence, evaluating their potential to improve image quality and tissue characterization. Conclusions: In conclusion, while technological evolution offers new diagnostic frontiers, a rigorous and systematic interpretation of CT findings remains the fundamental pillar for optimizing patient outcomes and reducing unnecessary interventions.
- 19 May 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (10) , 3915
Open AccessCT of Soft Tissue Infections: Current and Future Perspectives from Diagnosis to Treatment
Background/Objectives: Skin and soft-tissue infections represent a critical diagnostic challenge in the emergency department, requiring rapid differentiation between superficial processes and life-threatening necrotizing conditions. Methods: This clinical update analyzes the central role of CT in the assessment of these pathologies, proposing a management-oriented algorithm to support risk stratification and therapeutic planning. Results: By reviewing the fundamental semiological features of contrast-enhanced CT, this work highlights how the identification of critical markers directly guides the choice between medical management and urgent surgical intervention. In addition to consolidating protocols based on conventional CT, the article examines current perspectives and clinical limitations of emerging technologies, including Spectral CT (DECT and PCD-CT) and artificial intelligence, evaluating their potential to improve image quality and tissue characterization. Conclusions: In conclusion, while technological evolution offers new diagnostic frontiers, a rigorous and systematic interpretation of CT findings remains the fundamental pillar for optimizing patient outcomes and reducing unnecessary interventions.
- Open Access
- 24 April 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (9) , 3262
Radiomics Applied to the Diagnosis of Peripheral Nerve Disorders: A Systematic Review and Meta-Analysis of the Existing Literature
Background: This study aims to systematically review the current literature on the application of radiomic features and artificial intelligence (AI) in the diagnosis and prognosis of common peripheral nerve-related conditions, including carpal tunnel syndrome (CTS), chronic inflammatory demyelinating polyneuropathy (CIDP), polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy and skin abnormalities (POEMS) syndrome, and in distinguishing between benign and malignant tumors. Methods: A comprehensive literature search was conducted in PubMed and Google Scholar for studies published between January 2019 and September 2025. Inclusion criteria comprised studies that used radiomics or AI-based radiomics approaches with diagnostic or prognostic purposes in peripheral nerve disorders. Results: A total of 40 studies were identified, of which 17 met the inclusion criteria. Among these, 9 studies employed magnetic resonance imaging (MRI), including one combined with PET/CT, while 8 used ultrasound (US). Most studies were retrospective and limited by small sample sizes, lack of external validation, and predominance of single-center designs. Conclusions: Since a seminal study published in 2019, there has been increasing evidence supporting the role of radiomics and AI in improving the diagnosis and prognosis of peripheral nerve disorders, particularly using MRI and US. However, significant challenges remain, including variability in imaging data, segmentation complexity, and limited availability of validated datasets. Future advancements in imaging technologies and multidisciplinary collaboration are essential to enhance clinical applicability.
- 24 April 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (9) , 3262
Open AccessRadiomics Applied to the Diagnosis of Peripheral Nerve Disorders: A Systematic Review and Meta-Analysis of the Existing Literature
Background: This study aims to systematically review the current literature on the application of radiomic features and artificial intelligence (AI) in the diagnosis and prognosis of common peripheral nerve-related conditions, including carpal tunnel syndrome (CTS), chronic inflammatory demyelinating polyneuropathy (CIDP), polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy and skin abnormalities (POEMS) syndrome, and in distinguishing between benign and malignant tumors. Methods: A comprehensive literature search was conducted in PubMed and Google Scholar for studies published between January 2019 and September 2025. Inclusion criteria comprised studies that used radiomics or AI-based radiomics approaches with diagnostic or prognostic purposes in peripheral nerve disorders. Results: A total of 40 studies were identified, of which 17 met the inclusion criteria. Among these, 9 studies employed magnetic resonance imaging (MRI), including one combined with PET/CT, while 8 used ultrasound (US). Most studies were retrospective and limited by small sample sizes, lack of external validation, and predominance of single-center designs. Conclusions: Since a seminal study published in 2019, there has been increasing evidence supporting the role of radiomics and AI in improving the diagnosis and prognosis of peripheral nerve disorders, particularly using MRI and US. However, significant challenges remain, including variability in imaging data, segmentation complexity, and limited availability of validated datasets. Future advancements in imaging technologies and multidisciplinary collaboration are essential to enhance clinical applicability.
- Open Access
- 9 December 2025
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 14 (24) , 8712
Development of a Simplified Protocol for Respiratory Muscle Segmentation in Unenhanced Chest CT and Identification of New Radiomic Biomarkers of Sarcopenia in Lung Diseases: A Retrospective Study
Background/Objectives: Respiratory muscle sarcopenia worsens outcomes in chronic lung disease, and quantitative Computed Tomography (CT) may provide objective biomarkers; this study aimed to develop a time-efficient segmentation protocol and identify radiomic biomarkers of respiratory muscle sarcopenia. Methods: This retrospective study analyzed 30 unenhanced chest CT from adult patients. The whole volume of the pectoralis major (PM), pectoralis minor (Pm), serratus anterior (SA), and fourth intercostal (4I) muscles was manually segmented. Patients were classified as sarcopenic or non-sarcopenic. Radiomics features and mean muscle density were extracted using PyRadiomics. Features associated with sarcopenia were selected using Least Absolute Shrinkage and Selection Operator (LASSO) regression and backward stepwise selection. Four sets of slices consisting of one, three, five, and seven slices were then sampled from each muscle around a fixed anatomical landmark. Deviations of each set of slices from whole-muscle metrics were evaluated using Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Results: Features selection identified 25 biomarkers of sarcopenia in PM, 24 in Pm, and 34 in SA. Variability-related features were significantly associated with sarcopenia (OR = 2.26; p = 0.012), while structural features showed an inverse association (OR = 0.18; p = 0.004). Mean muscle density and most radiomic features were well represented by single slice for every muscle. In the PM and Pm, eight and six radiomic features were better approximated segmenting more than one slice (p < 0.05). Conclusions: Radiomics enables quantitative assessment of sarcopenia. For SA, a simplified segmentation protocol consisting of a single slice enables approximating muscle density and radiomics of whole muscle volume. For PM and Pm, three or more slices allow a better representation of 8 and 6 radiomic features, respectively.
- 9 December 2025
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 14 (24) , 8712
Open AccessDevelopment of a Simplified Protocol for Respiratory Muscle Segmentation in Unenhanced Chest CT and Identification of New Radiomic Biomarkers of Sarcopenia in Lung Diseases: A Retrospective Study
Background/Objectives: Respiratory muscle sarcopenia worsens outcomes in chronic lung disease, and quantitative Computed Tomography (CT) may provide objective biomarkers; this study aimed to develop a time-efficient segmentation protocol and identify radiomic biomarkers of respiratory muscle sarcopenia. Methods: This retrospective study analyzed 30 unenhanced chest CT from adult patients. The whole volume of the pectoralis major (PM), pectoralis minor (Pm), serratus anterior (SA), and fourth intercostal (4I) muscles was manually segmented. Patients were classified as sarcopenic or non-sarcopenic. Radiomics features and mean muscle density were extracted using PyRadiomics. Features associated with sarcopenia were selected using Least Absolute Shrinkage and Selection Operator (LASSO) regression and backward stepwise selection. Four sets of slices consisting of one, three, five, and seven slices were then sampled from each muscle around a fixed anatomical landmark. Deviations of each set of slices from whole-muscle metrics were evaluated using Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Results: Features selection identified 25 biomarkers of sarcopenia in PM, 24 in Pm, and 34 in SA. Variability-related features were significantly associated with sarcopenia (OR = 2.26; p = 0.012), while structural features showed an inverse association (OR = 0.18; p = 0.004). Mean muscle density and most radiomic features were well represented by single slice for every muscle. In the PM and Pm, eight and six radiomic features were better approximated segmenting more than one slice (p < 0.05). Conclusions: Radiomics enables quantitative assessment of sarcopenia. For SA, a simplified segmentation protocol consisting of a single slice enables approximating muscle density and radiomics of whole muscle volume. For PM and Pm, three or more slices allow a better representation of 8 and 6 radiomic features, respectively.
- Open Access
- 1 October 2025
- Published by MDPI AG in Preprints.org
Development of a Simplified Protocol for Respiratory Muscle Segmentation in Unenhanced Chest CT and Identification of New Radiomic Biomarkers of Sarcopenia in Lung Diseases: A Retrospective Study
Background/Objectives: Respiratory muscle sarcopenia worsens outcomes in chronic lung disease, and quantitative CT may provide objective biomarkers; this study aimed to develop a time-efficient segmentation protocol and identify radiomic biomarkers of respiratory muscle sarcopenia. Methods: This retrospective study analyzed 30 unenhanced chest CT from adult patients. The whole volume of the pectoralis major (PM), pectoralis minor (Pm), serratus anterior (SA), and fourth intercostal (4I) muscles were manually segmented. Patients were classified as sarcopenic or non-sarcopenic. Radiomics features and mean muscle density were extracted using PyRadiomics. Features associated with sarcopenia were selected using LASSO regression and backward stepwise selection. Four sets of slices consisting of one, three, five, and seven slices were then sampled from each muscle around a fixed anatomical landmark. Deviations of each set of slices from whole-muscle metrics were evaluated using MAE and MAPE. Results: Features selection identified 25 biomarkers of sarcopenia in PM, 24 in Pm, and 34 in SA. Variability-related features were significantly associated with sarcopenia (OR = 2.26; P = .012), while structural features showed an inverse association (OR = 0.18; P = .004). Mean muscle density and most radiomic features were well represented by single slice for every muscle. In the PM and Pm eight and six radiomic features were better approximated segmenting more than one slice (p < 0.05). Conclusions: Radiomics enables quantitative assessment of sarcopenia. For SA, a simplified segmentation protocol consisting of a single slice enables to approximate muscle density and radiomics of whole muscle volume. For PM and Pm, three or more slices allow a better representation of 8 and 6 radiomic features respectively.
- 1 October 2025
- Published by MDPI AG in Preprints.org
Open AccessDevelopment of a Simplified Protocol for Respiratory Muscle Segmentation in Unenhanced Chest CT and Identification of New Radiomic Biomarkers of Sarcopenia in Lung Diseases: A Retrospective Study
Background/Objectives: Respiratory muscle sarcopenia worsens outcomes in chronic lung disease, and quantitative CT may provide objective biomarkers; this study aimed to develop a time-efficient segmentation protocol and identify radiomic biomarkers of respiratory muscle sarcopenia. Methods: This retrospective study analyzed 30 unenhanced chest CT from adult patients. The whole volume of the pectoralis major (PM), pectoralis minor (Pm), serratus anterior (SA), and fourth intercostal (4I) muscles were manually segmented. Patients were classified as sarcopenic or non-sarcopenic. Radiomics features and mean muscle density were extracted using PyRadiomics. Features associated with sarcopenia were selected using LASSO regression and backward stepwise selection. Four sets of slices consisting of one, three, five, and seven slices were then sampled from each muscle around a fixed anatomical landmark. Deviations of each set of slices from whole-muscle metrics were evaluated using MAE and MAPE. Results: Features selection identified 25 biomarkers of sarcopenia in PM, 24 in Pm, and 34 in SA. Variability-related features were significantly associated with sarcopenia (OR = 2.26; P = .012), while structural features showed an inverse association (OR = 0.18; P = .004). Mean muscle density and most radiomic features were well represented by single slice for every muscle. In the PM and Pm eight and six radiomic features were better approximated segmenting more than one slice (p < 0.05). Conclusions: Radiomics enables quantitative assessment of sarcopenia. For SA, a simplified segmentation protocol consisting of a single slice enables to approximate muscle density and radiomics of whole muscle volume. For PM and Pm, three or more slices allow a better representation of 8 and 6 radiomic features respectively.