Lynn Farner
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Lynn Farner has a total of 55 co-authors. Here are the top co-authors by number of shared publications:
- Patrick Freund
Spinal Cord Injury Center, Balgrist University Hospital, University of Zurich, Zurich, Switzerland
7 shared publications - Dario Pfyffer
Spinal Cord Injury Center, Balgrist University Hospital, University of Zurich, Zurich, Switzerland
6 shared publications - Enamundram Naga Karthik
NeuroPoly Lab, Institute of Biomedical Engineering, Polytechnique Montréal, Montreal, QC, Canada
5 shared publications - Jan Valošek
NeuroPoly Lab, Institute of Biomedical Engineering, Polytechnique Montreal, Montreal, Canada
5 shared publications - Simon Schading-Sassenhausen
Spinal Cord Injury Center, Balgrist University Hospital University of Zurich Zurich Switzerland
5 shared publications - Andrew C. Smith
Physical Therapy Program, Department of Physical Medicine and Rehabilitation, School of Medicine, University of Colorado, Aurora, CO, USA
4 shared publications
Latest Publications
- Open Access
- 5 November 2026
- Published by Wiley in Annals of Clinical and Translational Neurology
Automatic Lesion Morphometrics Reliably Predict Sensorimotor Recovery After Spinal Cord Injury: A Longitudinal, Multicenter Neuroimaging Study
Objective: To evaluate automatic magnetic resonance imaging (MRI)‐derived lesion morphometrics against manual measurements and their prognostic value for characterizing distinct acute to chronic neurological recovery trajectories after spinal cord injury (SCI). Methods: We studied a retrospective, multicenter cohort of cervical and thoracolumbar SCI patients who underwent routine clinical MRI at 3–4 weeks post‐SCI and neurological examinations over 6 months. Lesion severity was quantified using both established manual measurements and a novel automatic approach. We assessed the between‐method agreement of MRI‐derived lesion morphometrics, including lesion length, lesion width, and tissue bridge width, and their ability to stratify patients according to 6‐month motor and sensory outcomes. Data‐driven prediction models were used to identify clinically meaningful lesion measures and their respective thresholds, differentiating subgroups of specific recovery profiles. Results: Eighty‐three patients from eight sites were included. Automatically derived lesion morphometrics showed moderate to strong agreement with established manual measures (Spearman's ρ: 0.540–0.821, p < 0.001), capturing inter‐individual differences in lesion severity and demonstrating comparable prognostic value. Lesion width represented the strongest predictor of 6‐month neurological outcomes, splitting patients into recovery‐specific subgroups. Patients with a smaller lesion width (threshold: 5.88–6.41 mm) demonstrated higher 6‐month sensorimotor scores and superior recovery trajectories, while tissue bridges (threshold: 0.17 mm) further refined motor prediction‐based stratification of more severely affected patients. Interpretation: Automatic MRI measures of lesion severity hold promise as scalable, time‐efficient in vivo biomarkers of SCI prognosis, with potential for clinical adoption to guide treatment decisions and multicenter trial planning.
- 5 November 2026
- Published by Wiley in Annals of Clinical and Translational Neurology
Open AccessAutomatic Lesion Morphometrics Reliably Predict Sensorimotor Recovery After Spinal Cord Injury: A Longitudinal, Multicenter Neuroimaging Study
Objective: To evaluate automatic magnetic resonance imaging (MRI)‐derived lesion morphometrics against manual measurements and their prognostic value for characterizing distinct acute to chronic neurological recovery trajectories after spinal cord injury (SCI). Methods: We studied a retrospective, multicenter cohort of cervical and thoracolumbar SCI patients who underwent routine clinical MRI at 3–4 weeks post‐SCI and neurological examinations over 6 months. Lesion severity was quantified using both established manual measurements and a novel automatic approach. We assessed the between‐method agreement of MRI‐derived lesion morphometrics, including lesion length, lesion width, and tissue bridge width, and their ability to stratify patients according to 6‐month motor and sensory outcomes. Data‐driven prediction models were used to identify clinically meaningful lesion measures and their respective thresholds, differentiating subgroups of specific recovery profiles. Results: Eighty‐three patients from eight sites were included. Automatically derived lesion morphometrics showed moderate to strong agreement with established manual measures (Spearman's ρ: 0.540–0.821, p < 0.001), capturing inter‐individual differences in lesion severity and demonstrating comparable prognostic value. Lesion width represented the strongest predictor of 6‐month neurological outcomes, splitting patients into recovery‐specific subgroups. Patients with a smaller lesion width (threshold: 5.88–6.41 mm) demonstrated higher 6‐month sensorimotor scores and superior recovery trajectories, while tissue bridges (threshold: 0.17 mm) further refined motor prediction‐based stratification of more severely affected patients. Interpretation: Automatic MRI measures of lesion severity hold promise as scalable, time‐efficient in vivo biomarkers of SCI prognosis, with potential for clinical adoption to guide treatment decisions and multicenter trial planning.
- Open Access
- 12 May 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Anti-Nogo-A NG101 treatment induces changes in spinal cord micro- and macrostructure following spinal cord injury
NG101 is a recombinant antibody that neutralizes the nerve growth inhibitor Nogo-A, promoting neural repair and improving upper extremity motor function in spinal cord injury (SCI). This study evaluated spinal cord MRI biomarkers to detect treatment-related structural changes and enhance patient stratification using data from 106 participants with acute cervical SCI in the phase 2b NISCI trial. We assessed lesion volume, tissue bridges, and remote changes in cross-sectional cord area (CSA), and tract-specific myelin-sensitive magnetization transfer saturation (MTsat) over six months. Compared to placebo, NG101-treated participants exhibited faster lesion volume reduction and a slower decline of CSA and MTsat in the corticospinal tracts and dorsal columns. Crucially, multimodal stratification incorporating MRI and electrophysiological measures substantially enhanced the detection of clinical treatment effects. These findings suggest NG101 slows trauma-induced progressive macro- and microstructural degeneration or promotes fiber sprouting. Combining MRI with electrophysiology enables sensitive detection of treatment effects and efficient trial designs. ClinicalTrials.gov identifier: NCT03935321.
- 12 May 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessAnti-Nogo-A NG101 treatment induces changes in spinal cord micro- and macrostructure following spinal cord injury
NG101 is a recombinant antibody that neutralizes the nerve growth inhibitor Nogo-A, promoting neural repair and improving upper extremity motor function in spinal cord injury (SCI). This study evaluated spinal cord MRI biomarkers to detect treatment-related structural changes and enhance patient stratification using data from 106 participants with acute cervical SCI in the phase 2b NISCI trial. We assessed lesion volume, tissue bridges, and remote changes in cross-sectional cord area (CSA), and tract-specific myelin-sensitive magnetization transfer saturation (MTsat) over six months. Compared to placebo, NG101-treated participants exhibited faster lesion volume reduction and a slower decline of CSA and MTsat in the corticospinal tracts and dorsal columns. Crucially, multimodal stratification incorporating MRI and electrophysiological measures substantially enhanced the detection of clinical treatment effects. These findings suggest NG101 slows trauma-induced progressive macro- and microstructural degeneration or promotes fiber sprouting. Combining MRI with electrophysiology enables sensitive detection of treatment effects and efficient trial designs. ClinicalTrials.gov identifier: NCT03935321.
- Open Access
Top-down processing alone activates the early somatosensory nuclei
The early somatosensory brainstem and thalamic nuclei are classically viewed as relays of peripheral input, yet research in animal models indicates they also receive top-down cortical signals. Whether such processing exists in humans remains unknown. Using cervical spinal cord injury (SCI), in which peripheral somatosensory input is reduced or absent while cortical processing is preserved, we tested whether top-down processing can elicit activation across the somatosensory nuclei. We combined 3 Tesla functional and quantitative MRI data to assess activity and structural properties along the somatosensory hand pathway in a cross-sectional study of 16 individuals with chronic cervical SCI (mean age ± s.e.m.=52.4 ± 3.5 years) and 20 age-, sex-, and handedness-matched able-bodied control subjects (mean age=50.8 ± 3.5 years). Participants were visually cued to make overt or, in cases of hand paralysis, attempted right- and left-hand movements. Activation was quantified across the cuneate nucleus, ventroposterior lateral thalamus, and primary somatosensory hand cortex, while structural properties were assessed using quantitative MRI measures sensitive to myelin and tissue integrity, including magnetisation transfer saturation (MTsat) and effective transverse relaxation rate (R2*), alongside morphometric measures. Despite reduced or absent peripheral input, SCI participants exhibited robust and lateralised activation across all levels of the somatosensory pathway. This pattern persisted even in a participant with complete hand paralysis who lacked bottom-up afferent input during the fMRI task, indicating that top-down processing alone is sufficient to drive activity in early somatosensory relays. We simultaneously observed structural degeneration in the cuneate nucleus of SCI participants, marked by reduced volume and myelin-sensitive metrics (MTsat and R2*), consistent with secondary degeneration. The extent of atrophy was related to time since injury and reduced sensorimotor hand function, but showed no significant relationship with functional activation, suggesting that preserved corticocuneate signalling is not dependent on the degree of structural degeneration. This provides the first evidence that the cuneate nuclei in humans are subject to both bottom- up and top-down somatosensory processing. Although these nuclei are vulnerable to structural atrophy following dorsal column injury, our results suggest that top-down processing remains intact decades after SCI. This may have implications for the development of rehabilitation treatments targeting preserved somatosensory processing after injury.
Open AccessTop-down processing alone activates the early somatosensory nuclei
The early somatosensory brainstem and thalamic nuclei are classically viewed as relays of peripheral input, yet research in animal models indicates they also receive top-down cortical signals. Whether such processing exists in humans remains unknown. Using cervical spinal cord injury (SCI), in which peripheral somatosensory input is reduced or absent while cortical processing is preserved, we tested whether top-down processing can elicit activation across the somatosensory nuclei. We combined 3 Tesla functional and quantitative MRI data to assess activity and structural properties along the somatosensory hand pathway in a cross-sectional study of 16 individuals with chronic cervical SCI (mean age ± s.e.m.=52.4 ± 3.5 years) and 20 age-, sex-, and handedness-matched able-bodied control subjects (mean age=50.8 ± 3.5 years). Participants were visually cued to make overt or, in cases of hand paralysis, attempted right- and left-hand movements. Activation was quantified across the cuneate nucleus, ventroposterior lateral thalamus, and primary somatosensory hand cortex, while structural properties were assessed using quantitative MRI measures sensitive to myelin and tissue integrity, including magnetisation transfer saturation (MTsat) and effective transverse relaxation rate (R2*), alongside morphometric measures. Despite reduced or absent peripheral input, SCI participants exhibited robust and lateralised activation across all levels of the somatosensory pathway. This pattern persisted even in a participant with complete hand paralysis who lacked bottom-up afferent input during the fMRI task, indicating that top-down processing alone is sufficient to drive activity in early somatosensory relays. We simultaneously observed structural degeneration in the cuneate nucleus of SCI participants, marked by reduced volume and myelin-sensitive metrics (MTsat and R2*), consistent with secondary degeneration. The extent of atrophy was related to time since injury and reduced sensorimotor hand function, but showed no significant relationship with functional activation, suggesting that preserved corticocuneate signalling is not dependent on the degree of structural degeneration. This provides the first evidence that the cuneate nuclei in humans are subject to both bottom- up and top-down somatosensory processing. Although these nuclei are vulnerable to structural atrophy following dorsal column injury, our results suggest that top-down processing remains intact decades after SCI. This may have implications for the development of rehabilitation treatments targeting preserved somatosensory processing after injury.
- Open Access
- 27 June 2025
- Published by Wiley in Annals of Clinical and Translational Neurology
- Vol. 12 (9) , 1762-1770
3D MRI Tract‐Specific Spinal Cord Lesion Pattern Improves Prediction of Distinct Neurological Recovery
Objective: To distinguish lateralized motor‐ and sensory‐tract damage after acute spinal cord injury (SCI) and explore its predictive power for motor and sensory recovery.Methods: Thirty‐five SCI patients (two female) from a multi‐center data set (placebo‐arm of the Nogo‐A‐Inhibition in SCI trial) underwent routine T2‐weighted sagittal MRI scans at the lesion site at baseline (19.9 days, 95% confidence interval [CI]: 17.9–21.8), 1‐month (54.2 days, 95% CI: 52.1–56.2), and 6‐month (192.4 days, 95% CI: 181.3–203.6) post‐injury. Concurrently with the MRI scans, clinical examinations were performed. Lesions were manually segmented across all slices, and 3D‐tract damage was assessed by determining the overlap between segmented lesions and identified motor and sensory tracts in the axial plane. The relationship between lesion assessments and baseline‐adjusted clinical outcomes at 6 months was explored.Results: Over the 6‐month, patients recovered by 4.95 motor points/month (95% CI: 3.89–5.89, p < 0.001) on the International Standards for the Neurological Classification of SCI scale, 2.28 light‐touch points/month (95% CI: 1.43–3.12, p < 0.001), and 2.06 pinprick points/month (95% CI: 1.21–2.91, p < 0.001). Lesion volume decreased from 381.82mm 3 (95% CI: 295.78–467.87) by −14.04 mm3/month (95% CI: −25.39 to −1.56, p = 0.023). MRI visible changes in motor tract damage over the 6‐month were marginal (0.02%/month, 95% CI: −0.81 to −1.02, p = 0.971). Changes in the sensory tracts were more pronounced, decreasing by −0.69%/month (95% CI: −1.29 to −0.09, p = 0.05). Left‐and‐right motor‐tract damage at baseline significantly predicted left‐and‐right motor score recovery (R2 = 0.75, p = 0.015), while baseline left‐and‐right sensory‐tract damage significantly predicted improvements in left‐and‐right pin‐prick scores (R2 = 0.79, p = 0.024).Interpretation: Revealing the extent of damage to spinal motor‐and sensory‐pathways early after SCI is a valuable predictor of related neurological recovery. Tracking 3D dynamics of major spinal pathways has the potential to enhance diagnostic accuracy and patient stratification for future clinical trials.
- 27 June 2025
- Published by Wiley in Annals of Clinical and Translational Neurology
- Vol. 12 (9) , 1762-1770
Open Access3D MRI Tract‐Specific Spinal Cord Lesion Pattern Improves Prediction of Distinct Neurological Recovery
Objective: To distinguish lateralized motor‐ and sensory‐tract damage after acute spinal cord injury (SCI) and explore its predictive power for motor and sensory recovery.Methods: Thirty‐five SCI patients (two female) from a multi‐center data set (placebo‐arm of the Nogo‐A‐Inhibition in SCI trial) underwent routine T2‐weighted sagittal MRI scans at the lesion site at baseline (19.9 days, 95% confidence interval [CI]: 17.9–21.8), 1‐month (54.2 days, 95% CI: 52.1–56.2), and 6‐month (192.4 days, 95% CI: 181.3–203.6) post‐injury. Concurrently with the MRI scans, clinical examinations were performed. Lesions were manually segmented across all slices, and 3D‐tract damage was assessed by determining the overlap between segmented lesions and identified motor and sensory tracts in the axial plane. The relationship between lesion assessments and baseline‐adjusted clinical outcomes at 6 months was explored.Results: Over the 6‐month, patients recovered by 4.95 motor points/month (95% CI: 3.89–5.89, p < 0.001) on the International Standards for the Neurological Classification of SCI scale, 2.28 light‐touch points/month (95% CI: 1.43–3.12, p < 0.001), and 2.06 pinprick points/month (95% CI: 1.21–2.91, p < 0.001). Lesion volume decreased from 381.82mm 3 (95% CI: 295.78–467.87) by −14.04 mm3/month (95% CI: −25.39 to −1.56, p = 0.023). MRI visible changes in motor tract damage over the 6‐month were marginal (0.02%/month, 95% CI: −0.81 to −1.02, p = 0.971). Changes in the sensory tracts were more pronounced, decreasing by −0.69%/month (95% CI: −1.29 to −0.09, p = 0.05). Left‐and‐right motor‐tract damage at baseline significantly predicted left‐and‐right motor score recovery (R2 = 0.75, p = 0.015), while baseline left‐and‐right sensory‐tract damage significantly predicted improvements in left‐and‐right pin‐prick scores (R2 = 0.79, p = 0.024).Interpretation: Revealing the extent of damage to spinal motor‐and sensory‐pathways early after SCI is a valuable predictor of related neurological recovery. Tracking 3D dynamics of major spinal pathways has the potential to enhance diagnostic accuracy and patient stratification for future clinical trials.
- 8 February 2025
- Published by Springer Nature
- p. 198-209
SCIsegV2: A Universal Tool for Segmentation of Intramedullary Lesions in Spinal Cord Injury
- 8 February 2025
- Published by Springer Nature
- p. 198-209
SCIsegV2: A Universal Tool for Segmentation of Intramedullary Lesions in Spinal Cord Injury
- 1 January 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (1) , e240005
SCIseg: Automatic Segmentation of Intramedullary Lesions in Spinal Cord Injury on T2-weighted MRI Scans
The proposed deep learning model accurately segmented the spinal cord and spinal cord injury lesions in a diverse, multicenter dataset of T2-weighted MRI scans.
- 1 January 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (1) , e240005
SCIseg: Automatic Segmentation of Intramedullary Lesions in Spinal Cord Injury on T2-weighted MRI Scans
The proposed deep learning model accurately segmented the spinal cord and spinal cord injury lesions in a diverse, multicenter dataset of T2-weighted MRI scans.
- Open Access
- 17 September 2024
- Published by SAGE Publications in Journal of Neurotrauma
- Vol. 41 (17-18) , 2146-2157
Extent of Traumatic Spinal Cord Injury Is Lesion Level Dependent and Predictive of Recovery: A Multicenter Neuroimaging Study
Assessing the extent of the intramedullary lesion after spinal cord injury (SCI) might help to improve prognostication. However, because the neurological level of injury impacts the recovery potential of SCI patients, the question arises whether lesion size parameters and predictive models based on those parameters are affected as well. In this retrospective observational study, the extent of the intramedullary lesion between individuals who sustained cervical and thoracolumbar SCI was compared, and its relation to clinical recovery was assessed. In total, 154 patients with subacute SCI (89 individuals with cervical lesions and 65 individuals with thoracolumbar lesions) underwent conventional clinical magnetic resonance imaging 1 month after injury and clinical examination at 1 and 12 months. The morphology of the focal lesion within the spinal cord was manually assessed on the midsagittal slice of T 2 -weighted magnetic resonance images and compared between cervical and thoracolumbar SCI patients, as well as between patients who improved at least one American Spinal Injury Association Impairment Scale (AIS) grade (converters) and patients without AIS grade improvement (nonconverters). The predictive value of lesion parameters including lesion length, lesion width, and preserved tissue bridges for predicting AIS grade conversion was assessed using regression models (conditional inference tree analysis). Lesion length was two times longer in thoracolumbar compared with cervical SCI patients (F = 39.48, p < 0.0001), whereas lesion width and tissue bridges width did not differ. When comparing AIS grade converters and nonconverters, converters showed a smaller lesion length (F = 5.46, p = 0.021), a smaller lesion width (F = 13.75, p = 0.0003), and greater tissue bridges (F = 12.87, p = 0.0005). Using regression models, tissue bridges allowed more refined subgrouping of patients in AIS groups B, C, and D according to individual recovery profiles between 1 month and 12 months after SCI, whereas lesion length added no additional information for further subgrouping. This study characterizes differences in the anteroposterior and craniocaudal lesion extents after SCI. The two times greater lesion length in thoracolumbar compared with cervical SCI might be related to differences in the anatomy, biomechanics, and perfusion between the cervical and thoracic spines. Preserved tissue bridges were less influenced by the lesion level while closely related to the clinical impairment. These results highlight the robustness and utility of tissue bridges as a neuroimaging biomarker for predicting the clinical outcome after SCI in heterogeneous patient populations and for patient stratification in clinical trials.
- 17 September 2024
- Published by SAGE Publications in Journal of Neurotrauma
- Vol. 41 (17-18) , 2146-2157
Open AccessExtent of Traumatic Spinal Cord Injury Is Lesion Level Dependent and Predictive of Recovery: A Multicenter Neuroimaging Study
Assessing the extent of the intramedullary lesion after spinal cord injury (SCI) might help to improve prognostication. However, because the neurological level of injury impacts the recovery potential of SCI patients, the question arises whether lesion size parameters and predictive models based on those parameters are affected as well. In this retrospective observational study, the extent of the intramedullary lesion between individuals who sustained cervical and thoracolumbar SCI was compared, and its relation to clinical recovery was assessed. In total, 154 patients with subacute SCI (89 individuals with cervical lesions and 65 individuals with thoracolumbar lesions) underwent conventional clinical magnetic resonance imaging 1 month after injury and clinical examination at 1 and 12 months. The morphology of the focal lesion within the spinal cord was manually assessed on the midsagittal slice of T 2 -weighted magnetic resonance images and compared between cervical and thoracolumbar SCI patients, as well as between patients who improved at least one American Spinal Injury Association Impairment Scale (AIS) grade (converters) and patients without AIS grade improvement (nonconverters). The predictive value of lesion parameters including lesion length, lesion width, and preserved tissue bridges for predicting AIS grade conversion was assessed using regression models (conditional inference tree analysis). Lesion length was two times longer in thoracolumbar compared with cervical SCI patients (F = 39.48, p < 0.0001), whereas lesion width and tissue bridges width did not differ. When comparing AIS grade converters and nonconverters, converters showed a smaller lesion length (F = 5.46, p = 0.021), a smaller lesion width (F = 13.75, p = 0.0003), and greater tissue bridges (F = 12.87, p = 0.0005). Using regression models, tissue bridges allowed more refined subgrouping of patients in AIS groups B, C, and D according to individual recovery profiles between 1 month and 12 months after SCI, whereas lesion length added no additional information for further subgrouping. This study characterizes differences in the anteroposterior and craniocaudal lesion extents after SCI. The two times greater lesion length in thoracolumbar compared with cervical SCI might be related to differences in the anatomy, biomechanics, and perfusion between the cervical and thoracic spines. Preserved tissue bridges were less influenced by the lesion level while closely related to the clinical impairment. These results highlight the robustness and utility of tissue bridges as a neuroimaging biomarker for predicting the clinical outcome after SCI in heterogeneous patient populations and for patient stratification in clinical trials.
- Open Access
SCIsegV2: A Universal Tool for Segmentation of Intramedullary Lesions in Spinal Cord Injury
Spinal cord injury (SCI) is a devastating incidence leading to permanent paralysis and loss of sensory-motor functions potentially resulting in the formation of lesions within the spinal cord. Imaging biomarkers obtained from magnetic resonance imaging (MRI) scans can predict the functional recovery of individuals with SCI and help choose the optimal treatment strategy. Currently, most studies employ manual quantification of these MRI-derived biomarkers, which is a subjective and tedious task. In this work, we propose (i) a universal tool for the automatic segmentation of intramedullary SCI lesions, dubbed \texttt{SCIsegV2}, and (ii) a method to automatically compute the width of the tissue bridges from the segmented lesion. Tissue bridges represent the spared spinal tissue adjacent to the lesion, which is associated with functional recovery in SCI patients. The tool was trained and validated on a heterogeneous dataset from 7 sites comprising patients from different SCI phases (acute, sub-acute, and chronic) and etiologies (traumatic SCI, ischemic SCI, and degenerative cervical myelopathy). Tissue bridges quantified automatically did not significantly differ from those computed manually, suggesting that the proposed automatic tool can be used to derive relevant MRI biomarkers. \texttt{SCIsegV2} and the automatic tissue bridges computation are open-source and available in Spinal Cord Toolbox (v6.4 and above) via the \texttt{sct\_deepseg -task seg\_sc\_lesion\_t2w\_sci} and \texttt{sct\_analyze\_lesion} functions, respectively.
Open AccessSCIsegV2: A Universal Tool for Segmentation of Intramedullary Lesions in Spinal Cord Injury
Spinal cord injury (SCI) is a devastating incidence leading to permanent paralysis and loss of sensory-motor functions potentially resulting in the formation of lesions within the spinal cord. Imaging biomarkers obtained from magnetic resonance imaging (MRI) scans can predict the functional recovery of individuals with SCI and help choose the optimal treatment strategy. Currently, most studies employ manual quantification of these MRI-derived biomarkers, which is a subjective and tedious task. In this work, we propose (i) a universal tool for the automatic segmentation of intramedullary SCI lesions, dubbed \texttt{SCIsegV2}, and (ii) a method to automatically compute the width of the tissue bridges from the segmented lesion. Tissue bridges represent the spared spinal tissue adjacent to the lesion, which is associated with functional recovery in SCI patients. The tool was trained and validated on a heterogeneous dataset from 7 sites comprising patients from different SCI phases (acute, sub-acute, and chronic) and etiologies (traumatic SCI, ischemic SCI, and degenerative cervical myelopathy). Tissue bridges quantified automatically did not significantly differ from those computed manually, suggesting that the proposed automatic tool can be used to derive relevant MRI biomarkers. \texttt{SCIsegV2} and the automatic tissue bridges computation are open-source and available in Spinal Cord Toolbox (v6.4 and above) via the \texttt{sct\_deepseg -task seg\_sc\_lesion\_t2w\_sci} and \texttt{sct\_analyze\_lesion} functions, respectively.
- Open Access
SCIseg: Automatic Segmentation of T2-weighted Intramedullary Lesions in Spinal Cord Injury
Background Quantitative MRI biomarkers in spinal cord injury (SCI) can help understand the extent of the focal injury. However, due to the lack of automatic segmentation methods, these biomarkers are derived manually, which is a time-consuming process prone to intra- and inter-rater variability, thus limiting large multi-site studies and translation to clinical workflows. Purpose To develop a deep learning tool for the automatic segmentation of T2-weighted hyperintense lesions and the spinal cord in SCI patients. Material and Methods This retrospective study included a cohort of SCI patients from three sites enrolled between July 2002 and February 2023 who underwent clinical MRI examination. A deep learning model, SCIseg, was trained on T2-weighted images with heterogeneous image resolutions (isotropic, anisotropic), and orientations (axial, sagittal) acquired using scanners from different manufacturers (Siemens, Philips, GE) and different field strengths (1T, 1.5T, 3T) for the automatic segmentation of SCI lesions and the spinal cord. The proposed method was visually and quantitatively compared with other open-source baseline methods. Quantitative biomarkers (lesion volume, lesion length, and maximal axial damage ratio) computed from manual ground-truth lesion masks and automatic SCIseg predictions were correlated with clinical scores (pinprick, light touch, and lower extremity motor scores). A between-group comparison was performed using the Wilcoxon signed-rank test. Results MRI data from 191 SCI patients (mean age, 48.1 years ± 17.9 [SD]; 142 males) were used for training. Compared to existing methods, SCIseg achieved the best segmentation performance for both the cord and lesions and generalized well to both traumatic and non-traumatic SCI patients. SCIseg is open-source and accessible through the Spinal Cord Toolbox. Conclusion Automatic segmentation of intramedullary lesions commonly seen in traumatic SCI replaces the tedious manual annotation process and enables the extraction of relevant lesion morphometrics in large cohorts. The proposed model generalizes across lesion etiologies (traumatic, ischemic), scanner manufacturers and heterogeneous image resolutions. Summary Automatic segmentation of the spinal cord and T2-weighted hyperintense lesions in spinal cord injury on MRI scans across different treatment strategies, lesion etiologies, sites, scanner manufacturers, and heterogeneous image resolutions. Key Results An open-source, automatic method, SCIseg, was trained on a dataset of 191 spinal cord injury patients from three sites for the segmentation of spinal cord and T2-weighted hyperintense lesions. SCIseg generalizes across traumatic and non-traumatic lesions, scanner manufacturers, and heterogeneous image resolutions, enabling the automatic extraction of lesion morphometrics in large multi-site cohorts. Morphometrics derived from the automatic predictions showed no statistically significant difference when compared with manual ground truth, implying reliability in SCIseg’s predictions.
Open AccessSCIseg: Automatic Segmentation of T2-weighted Intramedullary Lesions in Spinal Cord Injury
Background Quantitative MRI biomarkers in spinal cord injury (SCI) can help understand the extent of the focal injury. However, due to the lack of automatic segmentation methods, these biomarkers are derived manually, which is a time-consuming process prone to intra- and inter-rater variability, thus limiting large multi-site studies and translation to clinical workflows. Purpose To develop a deep learning tool for the automatic segmentation of T2-weighted hyperintense lesions and the spinal cord in SCI patients. Material and Methods This retrospective study included a cohort of SCI patients from three sites enrolled between July 2002 and February 2023 who underwent clinical MRI examination. A deep learning model, SCIseg, was trained on T2-weighted images with heterogeneous image resolutions (isotropic, anisotropic), and orientations (axial, sagittal) acquired using scanners from different manufacturers (Siemens, Philips, GE) and different field strengths (1T, 1.5T, 3T) for the automatic segmentation of SCI lesions and the spinal cord. The proposed method was visually and quantitatively compared with other open-source baseline methods. Quantitative biomarkers (lesion volume, lesion length, and maximal axial damage ratio) computed from manual ground-truth lesion masks and automatic SCIseg predictions were correlated with clinical scores (pinprick, light touch, and lower extremity motor scores). A between-group comparison was performed using the Wilcoxon signed-rank test. Results MRI data from 191 SCI patients (mean age, 48.1 years ± 17.9 [SD]; 142 males) were used for training. Compared to existing methods, SCIseg achieved the best segmentation performance for both the cord and lesions and generalized well to both traumatic and non-traumatic SCI patients. SCIseg is open-source and accessible through the Spinal Cord Toolbox. Conclusion Automatic segmentation of intramedullary lesions commonly seen in traumatic SCI replaces the tedious manual annotation process and enables the extraction of relevant lesion morphometrics in large cohorts. The proposed model generalizes across lesion etiologies (traumatic, ischemic), scanner manufacturers and heterogeneous image resolutions. Summary Automatic segmentation of the spinal cord and T2-weighted hyperintense lesions in spinal cord injury on MRI scans across different treatment strategies, lesion etiologies, sites, scanner manufacturers, and heterogeneous image resolutions. Key Results An open-source, automatic method, SCIseg, was trained on a dataset of 191 spinal cord injury patients from three sites for the segmentation of spinal cord and T2-weighted hyperintense lesions. SCIseg generalizes across traumatic and non-traumatic lesions, scanner manufacturers, and heterogeneous image resolutions, enabling the automatic extraction of lesion morphometrics in large multi-site cohorts. Morphometrics derived from the automatic predictions showed no statistically significant difference when compared with manual ground truth, implying reliability in SCIseg’s predictions.