Magnetic resonance imaging based anatomical assessment of tongue impairment due to amyotrophic lateral sclerosis: A preliminary study

AI-assisted automatic MRI-based tongue volume evaluation in motor neuron disease (MND)

PURPOSE

Motor neuron disease (MND) causes damage to the upper and lower motor neurons including the motor cranial nerves, the latter resulting in bulbar involvement with atrophy of the tongue muscle. To measure tongue atrophy, an operator independent automatic segmentation of the tongue is crucial. The aim of this study was to apply convolutional neural network (CNN) to MRI data in order to determine the volume of the tongue.

METHODS

A single triplanar CNN of U-Net architecture trained on axial, coronal, and sagittal planes was used for the segmentation of the tongue in MRI scans of the head. The 3D volumes were processed slice-wise across the three orientations and the predictions were merged using different voting strategies. This approach was developed using MRI datasets from 20 patients with 'classical' spinal amyotrophic lateral sclerosis (ALS) and 20 healthy controls and, in a pilot study, applied to the tongue volume quantification to 19 controls and 19 ALS patients with the variant progressive bulbar palsy (PBP).

RESULTS

Consensus models with softmax averaging and majority voting achieved highest segmentation accuracy and outperformed predictions on single orientations and consensus models with union and unanimous voting. At the group level, reduction in tongue volume was not observed in classical spinal ALS, but was significant in the PBP group, as compared to controls.

CONCLUSION

Utilizing single U-Net trained on three orthogonal orientations with consequent merging of respective orientations in an optimized consensus model reduces the number of erroneous detections and improves the segmentation of the tongue. The CNN-based automatic segmentation allows for accurate quantification of the tongue volumes in all subjects. The application to the ALS variant PBP showed significant reduction of the tongue volume in these patients and opens the way for unbiased future longitudinal studies in diseases affecting tongue volume.

Neurophysiological and imaging biomarkers of lower motor neuron dysfunction in motor neuron diseases/amyotrophic lateral sclerosis: IFCN handbook chapter

This chapter discusses comprehensive neurophysiological biomarkers utilised in motor neuron disease (MND) and, in particular, its commonest form, amyotrophic lateral sclerosis (ALS). These encompass the conventional techniques including nerve conduction studies (NCS), needle and high-density surface electromyography (EMG) and H-reflex studies as well as novel techniques. In the last two decades, new methods of assessing the loss of motor units in a muscle have been developed, that are more convenient than earlier methods of motor unit number estimation (MUNE),and may use either electrical stimulation (e.g. MScanFit MUNE) or voluntary activation (MUNIX). Electrical impedance myography (EIM) is another novel approach for the evaluation that relies upon the application and measurement of high-frequency, low-intensity electrical current. Nerve excitability techniques (NET) also provide insights into the function of an axon and reflect the changes in resting membrane potential, ion channel dysfunction and the structural integrity of the axon and myelin sheath. Furthermore, imaging ultrasound techniques as well as magnetic resonance imaging are capable of detecting the constituents of morphological changes in the nerve and muscle. The chapter provides a critical description of the ability of each technique to provide neurophysiological insight into the complex pathophysiology of MND/ALS. However, it is important to recognise the strengths and limitations of each approach in order to clarify utility. These neurophysiological biomarkers have demonstrated reliability, specificity and provide additional information to validate and assess lower motor neuron dysfunction. Their use has expanded the knowledge about MND/ALS and enhanced our understanding of the relationship between motor units, axons, reflexes and other neural circuits in relation to clinical features of patients with MND/ALS at different stages of the disease. Taken together, the ultimate goal is to aid early diagnosis, distinguish potential disease mimics, monitor and stage disease progression, quantify response to treatment and develop potential therapeutic interventions.

Ultrasonographic and manometric study of the tongue as biomarkers of dysphagia in patients with amyotrophic lateral sclerosis

BACKGROUND

The possibility of having methods to assess dysphagia in amyotrophic lateral sclerosis (ALS) patients in a minimally invasive manner could facilitate follow-up and allow performing of therapeutic interventions at earlier stages of the disease. The aim of the study was to analyze the role of tongue strength and thickness in ALS patients and their correlation with dysphagia and bulbar function.

METHODS

A sample of outpatients with ALS was evaluated for demographic and clinical features. Tongue thickness and strength have been measured for each patient, and quantitative and qualitative data of the videofluoroscopy swallow study have been analyzed.

RESULTS

Of the 38 ALS patients studied, 47.4% were women, and 26.3% had bulbar onset. The median time between symptom onset and the study was 24 months (IQR 11.5-48), and 55.3% of the patients were carriers of non-invasive mechanical ventilation. Tongue strength identified patients with impaired oral and pharyngeal transit and those with bolus residue scale (BRS) > 1 or penetration-aspiration scale (PAS) ≥ 3. In contrast, tongue thickness is only associated with impaired oral transit. Finally, anterior tongue strength ≤ 34 kPa and posterior tongue strength ≤ 34.5 kPa detected ALS penetrators/aspirators (PAS ≥ 3) and patients with ALS with post-swallow residue (BRS > 1).

CONCLUSIONS

Our results suggest that measures that assess the functionality (strength) of the tongue are more valuable than morphological measurements (thickness) for the follow-up of patients with ALS. Alterations of the anterior and posterior lingual strength correlate with the presence of bronchoaspiration and post-swallowing residue (BRS > 1).

VoxelHop: Successive Subspace Learning for ALS Disease Classification Using Structural MRI

Deep learning has great potential for accurate detection and classification of diseases with medical imaging data, but the performance is often limited by the number of training datasets and memory requirements. In addition, many deep learning models are considered a "black-box," thereby often limiting their adoption in clinical applications. To address this, we present a successive subspace learning model, termed VoxelHop, for accurate classification of Amyotrophic Lateral Sclerosis (ALS) using T2-weighted structural MRI data. Compared with popular convolutional neural network (CNN) architectures, VoxelHop has modular and transparent structures with fewer parameters without any backpropagation, so it is well-suited to small dataset size and 3D imaging data. Our VoxelHop has four key components, including (1) sequential expansion of near-to-far neighborhood for multi-channel 3D data; (2) subspace approximation for unsupervised dimension reduction; (3) label-assisted regression for supervised dimension reduction; and (4) concatenation of features and classification between controls and patients. Our experimental results demonstrate that our framework using a total of 20 controls and 26 patients achieves an accuracy of 93.48 % and an AUC score of 0.9394 in differentiating patients from controls, even with a relatively small number of datasets, showing its robustness and effectiveness. Our thorough evaluations also show its validity and superiority to the state-of-the-art 3D CNN classification approaches. Our framework can easily be generalized to other classification tasks using different imaging modalities.

Measuring Strain in Diffusion-Weighted Data Using Tagged Magnetic Resonance Imaging

Accurate strain measurement in a deforming organ has been essential in motion analysis using medical images. In recent years, internal tissue's in vivo motion and strain computation has been mostly achieved through dynamic magnetic resonance (MR) imaging. However, such data lack information on tissue's intrinsic fiber directions, preventing computed strain tensors from being projected onto a direction of interest. Although diffusion-weighted MR imaging excels at providing fiber tractography, it yields static images unmatched with dynamic MR data. This work reports an algorithm workflow that estimates strain values in the diffusion MR space by matching corresponding tagged dynamic MR images. We focus on processing a dataset of various human tongue deformations in speech. The geometry of tongue muscle fibers is provided by diffusion tractography, while spatiotemporal motion fields are provided by tagged MR analysis. The tongue's deforming shapes are determined by segmenting a synthetic cine dynamic MR sequence generated from tagged data using a deep neural network. Estimated motion fields are transformed into the diffusion MR space using diffeomorphic registration, eventually leading to strain values computed in the direction of muscle fibers. The method was tested on 78 time volumes acquired during three sets of specific tongue deformations including both speech and protrusion motion. Strain in the line of action of seven internal tongue muscles was extracted and compared both intra- and inter-subject. Resulting compression and stretching patterns of individual muscles revealed the unique behavior of individual muscles and their potential activation pattern.

A deep joint sparse non-negative matrix factorization framework for identifying the common and subject-specific functional units of tongue motion during speech

Intelligible speech is produced by creating varying internal local muscle groupings-i.e., functional units-that are generated in a systematic and coordinated manner. There are two major challenges in characterizing and analyzing functional units. First, due to the complex and convoluted nature of tongue structure and function, it is of great importance to develop a method that can accurately decode complex muscle coordination patterns during speech. Second, it is challenging to keep identified functional units across subjects comparable due to their substantial variability. In this work, to address these challenges, we develop a new deep learning framework to identify common and subject-specific functional units of tongue motion during speech. Our framework hinges on joint deep graph-regularized sparse non-negative matrix factorization (NMF) using motion quantities derived from displacements by tagged Magnetic Resonance Imaging. More specifically, we transform NMF with sparse and graph regularizations into modular architectures akin to deep neural networks by means of unfolding the Iterative Shrinkage-Thresholding Algorithm to learn interpretable building blocks and associated weighting map. We then apply spectral clustering to common and subject-specific weighting maps from which we jointly determine the common and subject-specific functional units. Experiments carried out with simulated datasets show that the proposed method achieved on par or better clustering performance over the comparison methods.Experiments carried out with in vivo tongue motion data show that the proposed method can determine the common and subject-specific functional units with increased interpretability and decreased size variability.

Muscle MRI in motor neuron diseases: a systematic review

Objective: To summarize applications of muscle magnetic resonance imaging (MRI) in cross-sectional assessment and longitudinal monitoring of motor neuron diseases and evaluate associations with clinical assessment techniques.Methods: PubMed and Scopus were searched for research published up to May 2021 relating to muscle MRI in motor neuron diseases, according to predefined inclusion and exclusion criteria. Studies were systematically appraised for bias and data were extracted for discussion.Results: Twenty-eight papers met inclusion criteria. The studies assessed muscle T1- and T2-weighted signal, diffusion, muscle volume, and fat infiltration, employing quantitative, qualitative, and semi-quantitative approaches. Various regions of interest were considered; changes in thigh and calf muscles were most frequently reported. Preliminary evidence of concordance between clinical and radiological findings and utility as an objective longitudinal biomarker is emerging.Conclusion: Muscle MRI appears a promising objective, versatile, and practical biomarker to assess motor neuron diseases.

Quantitative ultrasound of the tongue: Echo intensity is a potential biomarker of bulbar dysfunction in amyotrophic lateral sclerosis

OBJECTIVES

To learn if quantitative ultrasound (QUS) distinguishes the tongues of healthy participants and amyotrophic lateral sclerosis (ALS) patients by echo intensity (EI) and to evaluate if EI correlates with measures of bulbar function.

METHODS

Ultrasound was performed along the midline of the anterior tongue surface in 16 ALS patients and 16 age-matched controls using a linear hockey stick 16-7 MHz transducer. A region of interest was manually drawn and then EI was determined for the upper 1/3 of the muscle. For patients, the ALS functional rating scale - revised (ALSFRS-R) was used to calculate bulbar sub-scores and the Iowa Oral Performance Instrument (IOPI) was used to measure tongue strength.

RESULTS

EI was significantly higher in ALS patients than in healthy participants (49.8 versus 37.8 arbitrary units, p < 0.01). In the patient group, EI was negatively correlated with ALSFRS-R bulbar sub-score (R_S = -0.65, p < 0.01). An inverse correlation between EI and tongue strength did not reach significance (R_S = -0.34, p = 0.28).

CONCLUSIONS

This study suggests that EI can differentiate healthy from diseased tongue muscle, and correlates with a standard functional measure in ALS patients.

SIGNIFICANCE

Tongue EI may represent a novel biomarker for bulbar dysfunction in ALS.

Sonographic and 3T-MRI-based evaluation of the tongue in ALS

A few systematic imaging studies employing ultrasound (HRUS) and magnetic resonance imaging (MRI) have suggested tongue measures to aid in diagnosis of amyotrophic lateral sclerosis (ALS). The relationship between structural tongue alterations and the ALS patients' bulbar and overall motor function has not yet been elucidated. We here thus aimed to understand how in-vivo tongue alterations relate to motor function and motor function evolution over time in ALS. Our study included 206 ALS patients and 104 age- and sex-matched controls that underwent HRUS and 3T MRI of the tongue at baseline. Sonographic measures comprised coronal tongue echointensity, area, height, width and height/width ratio, while MRI measures comprised sagittal T1 intensity, tongue area, position and shape. Imaging-derived markers were related to baseline and longitudinal bulbar and overall motor function. Baseline T1 intensity was lower in ALS patients with more severe bulbar involvement at baseline. Smaller baseline coronal (HRUS) and sagittal (MRI) tongue area, smaller coronal height (HRUS) and width (HRUS) as well as more rounded sagittal tongue shape predicated more rapid functional impairment - not only of bulbar, but also of overall motor function - in ALS. Our results suggest that in-vivo sonography und MRI tongue measures could aid as biomarkers to reflect bulbar and motor function impairment.

Longitudinal multi-modal muscle-based biomarker assessment in motor neuron disease

Background

Clinical phenotypic heterogeneity represents a major barrier to trials in motor neuron disease (MND) and objective surrogate outcome measures are required, especially for slowly progressive patients. We assessed responsiveness of clinical, electrophysiological and radiological muscle-based assessments to detect MND-related progression.

Materials and methods

A prospective, longitudinal cohort study of 29 MND patients and 22 healthy controls was performed. Clinical measures, electrophysiological motor unit number index/size (MUNIX/MUSIX) and relative T2- and diffusion-weighted whole-body muscle magnetic resonance (MR) were assessed three times over 12 months. Multi-variable regression models assessed between-group differences, clinico-electrophysiological associations, and longitudinal changes. Standardized response means (SRMs) assessed sensitivity to change over 12 months.

Results

MND patients exhibited 18% higher whole-body mean muscle relative T2-signal than controls (95% CI 7–29%, p < 0.01), maximal in leg muscles (left tibialis anterior 71% (95% CI 33–122%, p < 0.01). Clinical and electrophysiological associations were evident. By 12 months, 16 patients had died or could not continue. In the remainder, relative T2-signal increased over 12 months by 14–29% in right tibialis anterior, right quadriceps, bilateral hamstrings and gastrocnemius/soleus (p < 0.01), independent of onset-site, and paralleled progressive weakness and electrophysiological loss of motor units. Highest clinical, electrophysiological and radiological SRMs were found for revised ALS-functional rating scale scores (1.22), tibialis anterior MUNIX (1.59), and relative T2-weighted leg muscle MR (right hamstrings: 0.98), respectively. Diffusion MR detected minimal changes.

Conclusion

MUNIX and relative T2-weighted MR represent objective surrogate markers of progressive denervation in MND. Radiological changes were maximal in leg muscles, irrespective of clinical onset-site.

Electronic supplementary material

The online version of this article (10.1007/s00415-019-09580-x) contains supplementary material, which is available to authorized users.

Skeletal muscle MRI differentiates SBMA and ALS and correlates with disease severity

OBJECTIVE

To investigate the use of muscle MRI for the differential diagnosis and as a disease progression biomarker for 2 major forms of motor neuron disorders: spinal bulbar muscular atrophy (SBMA) and amyotrophic lateral sclerosis (ALS).

METHODS

We applied quantitative 3-point Dixon and semiquantitative T1-weighted and short tau inversion recovery (STIR) imaging to bulbar and lower limb muscles and performed clinical and functional assessments in ALS (n = 21) and SBMA (n = 21), alongside healthy controls (n = 16). Acquired images were analyzed for the presence of fat infiltration or edema as well as specific patterns of muscle involvement. Quantitative MRI measurements were correlated with clinical measures of disease severity in ALS and SBMA.

RESULTS

Quantitative imaging revealed significant fat infiltration in bulbar (p < 0.001) and limb muscles in SBMA compared to controls (thigh: p < 0.001; calf: p = 0.001), identifying a characteristic pattern of muscle involvement. In ALS, semiquantitative STIR imaging detected marked hyperintensities in lower limb muscles, distinguishing ALS from SBMA and controls. Finally, MRI measurements correlated significantly with clinical scales of disease severity in both ALS and SBMA.

CONCLUSIONS

Our findings show that muscle MRI differentiates between SBMA and ALS and correlates with disease severity, supporting its use as a diagnostic tool and biomarker for disease progression. This highlights the clinical utility of muscle MRI in motor neuron disorders and contributes to establish objective outcome measures, which is crucial for the development of new drugs.

Intermittently tagged real-time MRI reveals internal tongue motion during speech production

PURPOSE

To demonstrate a tagging method compatible with RT-MRI for the study of speech production.

METHODS

Tagging is applied as a brief interruption to a continuous real-time spiral acquisition. Tagging can be initiated manually by the operator, cued to the speech stimulus, or be automatically applied with a fixed frequency. We use a standard 2D 1-3-3-1 binomial SPAtial Modulation of Magnetization (SPAMM) sequence with 1 cm spacing in both in-plane directions. Tag persistence in tongue muscle is simulated and validated in vivo. The ability to capture internal tongue deformations is tested during speech production of American English diphthongs in native speakers.

RESULTS

We achieved an imaging window of 650-800 ms at 1.5T, with imaging signal to noise ratio ≥ 17 and tag contrast to noise ratio ≥ 5 in human tongue, providing 36 frames/s temporal resolution and 2 mm in-plane spatial resolution with real-time interactive acquisition and view-sharing reconstruction. The proposed method was able to capture tongue motion patterns and their relative timing with adequate spatiotemporal resolution during the production of American English diphthongs and consonants.

CONCLUSION

Intermittent tagging during real-time MRI of speech production is able to reveal the internal deformations of the tongue. This capability will allow new investigations of valuable spatiotemporal information on the biomechanics of the lingual subsystems during speech without reliance on binning speech utterance repetition.

A Sparse Non-Negative Matrix Factorization Framework for Identifying Functional Units of Tongue Behavior From MRI

Muscle coordination patterns of lingual behaviors are synergies generated by deforming local muscle groups in a variety of ways. Functional units are functional muscle groups of local structural elements within the tongue that compress, expand, and move in a cohesive and consistent manner. Identifying the functional units using tagged-magnetic resonance imaging (MRI) sheds light on the mechanisms of normal and pathological muscle coordination patterns, yielding improvement in surgical planning, treatment, or rehabilitation procedures. In this paper, to mine this information, we propose a matrix factorization and probabilistic graphical model framework to produce building blocks and their associated weighting map using motion quantities extracted from tagged-MRI. Our tagged-MRI imaging and accurate voxel-level tracking provide previously unavailable internal tongue motion patterns, thus revealing the inner workings of the tongue during speech or other lingual behaviors. We then employ spectral clustering on the weighting map to identify the cohesive regions defined by the tongue motion that may involve multiple or undocumented regions. To evaluate our method, we perform a series of experiments. We first use two-dimensional images and synthetic data to demonstrate the accuracy of our method. We then use three-dimensional synthetic and in vivo tongue motion data using protrusion and simple speech tasks to identify subject-specific and data-driven functional units of the tongue in localized regions.