scholarly journals 3D Automated Segmentation of Lower Leg Muscles Using Machine Learning on a Heterogeneous Dataset

Diagnostics ◽  
2021 ◽  
Vol 11 (10) ◽  
pp. 1747
Author(s):  
Marlena Rohm ◽  
Marius Markmann ◽  
Johannes Forsting ◽  
Robert Rehmann ◽  
Martijn Froeling ◽  
...  
Keyword(s):  
Machine Learning ◽  
Test Performance ◽  
Imaging Techniques ◽  
Neuromuscular Diseases ◽  
Lower Leg ◽  
Heterogeneous Data ◽  
Quantitative Mri ◽  
Automated Segmentation ◽  
Heterogeneous Database ◽  
Leg Muscles

Quantitative MRI combines non-invasive imaging techniques to reveal alterations in muscle pathophysiology. Creating muscle-specific labels manually is time consuming and requires an experienced examiner. Semi-automatic and fully automatic methods reduce segmentation time significantly. Current machine learning solutions are commonly trained on data from healthy subjects using homogeneous databases with the same image contrast. While yielding high Dice scores (DS), those solutions are not applicable to different image contrasts and acquisitions. Therefore, the aim of our study was to evaluate the feasibility of automatic segmentation of a heterogeneous database. To create a heterogeneous dataset, we pooled lower leg muscle images from different studies with different contrasts and fields-of-view, containing healthy controls and diagnosed patients with various neuromuscular diseases. A second homogenous database with uniform contrasts was created as a subset of the first database. We trained three 3D-convolutional neuronal networks (CNN) on those databases to test performance as compared to manual segmentation. All networks, training on heterogeneous data, were able to predict seven muscles with a minimum average DS of 0.75. U-Net performed best when trained on the heterogeneous dataset (DS: 0.80 ± 0.10, AHD: 0.39 ± 0.35). ResNet and DenseNet yielded higher DS, when trained on a heterogeneous dataset (both DS: 0.86), as compared to a homogeneous dataset (ResNet DS: 0.83, DenseNet DS: 0.76). In conclusion, a CNN trained on a heterogeneous dataset achieves more accurate labels for predicting a heterogeneous database of lower leg muscles than a CNN trained on a homogenous dataset. We propose that a large heterogeneous database is needed, to make automated segmentation feasible for different kinds of image acquisitions.

scholarly journals Lower extremity muscle pathology in myotonic dystrophy type 1 assessed by quantitative MRI

Neurology ◽  
2019 ◽  
Vol 92 (24) ◽  
pp. e2803-e2814 ◽  
Author(s):  
Linda Heskamp ◽  
Marlies van Nimwegen ◽  
Marieke J. Ploegmakers ◽  
Guillaume Bassez ◽  
Jean-Francois Deux ◽  
...  
Keyword(s):  
Lower Extremity ◽  
Myotonic Dystrophy ◽  
Lower Leg ◽  
Muscle Volume ◽  
Quantitative Mri ◽  
Myotonic Dystrophy Type 1 ◽  
Myotonic Dystrophy Type ◽  
Leg Muscles ◽  
Fat Fraction

ObjectiveTo determine the value of quantitative MRI in providing imaging biomarkers for disease in 20 different upper and lower leg muscles of patients with myotonic dystrophy type 1 (DM1).MethodsWe acquired images covering these muscles in 33 genetically and clinically well-characterized patients with DM1 and 10 unaffected controls. MRIs were recorded with a Dixon method to determine muscle fat fraction, muscle volume, and contractile muscle volume, and a multi-echo spin-echo sequence was used to determine T2 water relaxation time (T2water), reflecting putative edema.ResultsMuscles in patients with DM1 had higher fat fractions than muscles of controls (15.6 ± 11.1% vs 3.7 ± 1.5%). In addition, patients had smaller muscle volumes (902 ± 232 vs 1,097 ± 251 cm3), smaller contractile muscle volumes (779 ± 247 vs 1,054 ± 246 cm3), and increased T2water (33.4 ± 1.0 vs 31.9 ± 0.6 milliseconds), indicating atrophy and edema, respectively. Lower leg muscles were affected most frequently, especially the gastrocnemius medialis and soleus. Distribution of fat content per muscle indicated gradual fat infiltration in DM1. Between-patient variation in fat fraction was explained by age (≈45%), and another ≈14% was explained by estimated progenitor CTG repeat length (r2 = 0.485) and somatic instability (r2 = 0.590). Fat fraction correlated with the 6-minute walk test (r = −0.553) and muscular impairment rating scale (r = 0.537) and revealed subclinical muscle involvement.ConclusionThis cross-sectional quantitative MRI study of 20 different lower extremity muscles in patients with DM1 revealed abnormal values for muscle fat fraction, volume, and T2water, which therefore may serve as objective biomarkers to assess disease state of skeletal muscles in these patients.ClinicalTrials.gov identifierNCT02118779.

A Review of Various Machine Learning Techniques for Brain Tumor Detection from MRI Images

2020 ◽  
Vol 16 (8) ◽  
pp. 937-945
Author(s):  
Aaishwarya Sanjay Bajaj ◽  
Usha Chouhan
Keyword(s):  
Machine Learning ◽  
Brain Tumor ◽  
Ct Scan ◽  
Imaging Techniques ◽  
Critical Evaluation ◽  
Tumor Detection ◽  
Machine Learning Techniques ◽  
X Rays ◽  
Detection Techniques ◽  
The Brain

Background: This paper endeavors to identify an expedient approach for the detection of the brain tumor in MRI images. The detection of tumor is based on i) review of the machine learning approach for the identification of brain tumor and ii) review of a suitable approach for brain tumor detection. Discussion: This review focuses on different imaging techniques such as X-rays, PET, CT- Scan, and MRI. This survey identifies a different approach with better accuracy for tumor detection. This further includes the image processing method. In most applications, machine learning shows better performance than manual segmentation of the brain tumors from MRI images as it is a difficult and time-consuming task. For fast and better computational results, radiology used a different approach with MRI, CT-scan, X-ray, and PET. Furthermore, summarizing the literature, this paper also provides a critical evaluation of the surveyed literature which reveals new facets of research. Conclusion: The problem faced by the researchers during brain tumor detection techniques and machine learning applications for clinical settings have also been discussed.

scholarly journals Supervised Machine Learning Methods and Hyperspectral Imaging Techniques Jointly Applied for Brain Cancer Classification

Sensors ◽  
2021 ◽  
Vol 21 (11) ◽  
pp. 3827
Author(s):  
Gemma Urbanos ◽  
Alberto Martín ◽  
Guillermo Vázquez ◽  
Marta Villanueva ◽  
Manuel Villa ◽  
...  
Keyword(s):  
Machine Learning ◽  
Blood Vessel ◽  
Hyperspectral Imaging ◽  
Imaging Techniques ◽  
Venous Blood ◽  
Healthy Tissue ◽  
Supervised Machine Learning ◽  
Support Vector ◽  
Arterial Blood

Hyperspectral imaging techniques (HSI) do not require contact with patients and are non-ionizing as well as non-invasive. As a consequence, they have been extensively applied in the medical field. HSI is being combined with machine learning (ML) processes to obtain models to assist in diagnosis. In particular, the combination of these techniques has proven to be a reliable aid in the differentiation of healthy and tumor tissue during brain tumor surgery. ML algorithms such as support vector machine (SVM), random forest (RF) and convolutional neural networks (CNN) are used to make predictions and provide in-vivo visualizations that may assist neurosurgeons in being more precise, hence reducing damages to healthy tissue. In this work, thirteen in-vivo hyperspectral images from twelve different patients with high-grade gliomas (grade III and IV) have been selected to train SVM, RF and CNN classifiers. Five different classes have been defined during the experiments: healthy tissue, tumor, venous blood vessel, arterial blood vessel and dura mater. Overall accuracy (OACC) results vary from 60% to 95% depending on the training conditions. Finally, as far as the contribution of each band to the OACC is concerned, the results obtained in this work are 3.81 times greater than those reported in the literature.

scholarly journals Radiomics: a new tool to differentiate adrenocortical adenoma from carcinoma

BJS Open ◽  
2021 ◽  
Vol 5 (1) ◽  
Author(s):  
F Torresan ◽  
F Crimì ◽  
F Ceccato ◽  
F Zavan ◽  
M Barbot ◽  
...  
Keyword(s):  
Machine Learning ◽  
Imaging Techniques ◽  
Second Order ◽  
Learning Approach ◽  
Adrenocortical Adenoma ◽  
Unsupervised Machine Learning ◽  
First Order ◽  
Main Challenge ◽  
Machine Learning Approach ◽  
Adrenocortical Tumours

Abstract Background The main challenge in the management of indeterminate incidentally discovered adrenal tumours is to differentiate benign from malignant lesions. In the absence of clear signs of invasion or metastases, imaging techniques do not always precisely define the nature of the mass. The present pilot study aimed to determine whether radiomics may predict malignancy in adrenocortical tumours. Methods CT images in unenhanced, arterial, and venous phases from 19 patients who had undergone resection of adrenocortical tumours and a cohort who had undergone surveillance for at least 5 years for incidentalomas were reviewed. A volume of interest was drawn for each lesion using dedicated software, and, for each phase, first-order (histogram) and second-order (grey-level colour matrix and run-length matrix) radiological features were extracted. Data were revised by an unsupervised machine learning approach using the K-means clustering technique. Results Of operated patients, nine had non-functional adenoma and 10 carcinoma. There were 11 patients in the surveillance group. Two first-order features in unenhanced CT and one in arterial CT, and 14 second-order parameters in unenhanced and venous CT and 10 second-order features in arterial CT, were able to differentiate adrenocortical carcinoma from adenoma (P < 0.050). After excluding two malignant outliers, the unsupervised machine learning approach correctly predicted malignancy in seven of eight adrenocortical carcinomas in all phases. Conclusion Radiomics with CT texture analysis was able to discriminate malignant from benign adrenocortical tumours, even by an unsupervised machine learning approach, in nearly all patients.

Asymmetrical Neural Adaptation in Lower Leg Muscles as a Consequence of Stereotypical Motor Training

2012 ◽  
Vol 44 (1) ◽  
pp. 63-68 ◽  
Author(s):  
Tetsuya Ogawa ◽  
Noritaka Kawashima ◽  
Shuji Suzuki ◽  
Kimitaka Nakazawa
Keyword(s):  
Lower Leg ◽  
Neural Adaptation ◽  
Motor Training ◽  
Leg Muscles

scholarly journals Entity Type Recognition for Heterogeneous Semantic Graphs

AI Magazine ◽  
2015 ◽  
Vol 36 (1) ◽  
pp. 75-86 ◽  
Author(s):  
Jennifer Sleeman ◽  
Tim Finin ◽  
Anupam Joshi
Keyword(s):  
Machine Learning ◽  
Background Knowledge ◽  
Knowledge Bases ◽  
Heterogeneous Data ◽  
Unstructured Data ◽  
Supervised Machine Learning ◽  
Coreference Resolution ◽  
Multiple Sources ◽  
Fine Grained ◽  
High Level

We describe an approach for identifying fine-grained entity types in heterogeneous data graphs that is effective for unstructured data or when the underlying ontologies or semantic schemas are unknown. Identifying fine-grained entity types, rather than a few high-level types, supports coreference resolution in heterogeneous graphs by reducing the number of possible coreference relations that must be considered. Big data problems that involve integrating data from multiple sources can benefit from our approach when the datas ontologies are unknown, inaccessible or semantically trivial. For such cases, we use supervised machine learning to map entity attributes and relations to a known set of attributes and relations from appropriate background knowledge bases to predict instance entity types. We evaluated this approach in experiments on data from DBpedia, Freebase, and Arnetminer using DBpedia as the background knowledge base.

Long-Lasting Changes in Muscle Twitch Force During Simulated Work While Standing or Walking

2016 ◽  
Vol 58 (8) ◽  
pp. 1117-1127 ◽  
Author(s):  
Maria-Gabriela Garcia ◽  
Rudolf Wall ◽  
Benjamin Steinhilber ◽  
Thomas Läubli ◽  
Bernard J. Martin
Keyword(s):  
Muscle Function ◽  
Signal Amplitude ◽  
Lower Leg ◽  
Muscle Twitch ◽  
Prolonged Standing ◽  
Twitch Force ◽  
Slow Pace ◽  
Leg Muscles ◽  
Ergonomic Interventions ◽  
Hard Floor

Objective: The aim of this study was to evaluate the long-lasting effects of prolonged standing work on a hard floor or floor mat and slow-pace walking on muscle twitch force (MTF) elicited by electrical stimulation. Background: Prolonged standing work may alter lower-leg muscle function, which can be quantified by changes in the MTF amplitude and duration related to muscle fatigue. Ergonomic interventions have been proposed to mitigate fatigue and discomfort; however, their influences remain controversial. Method: Ten men and eight women simulated standing work in 320-min experiments with three conditions: standing on a hard floor or an antifatigue mat and walking on a treadmill, each including three seated rest breaks. MTF in the gastrocnemius-soleus muscles was evaluated through changes in signal amplitude and duration. Results: The significant decrease of MTF amplitude and an increase of duration after standing work on a hard floor and on a mat persisted beyond 1 hr postwork. During walking, significant MTF metrics changes appeared 30 min postwork. MTF amplitude decrease was not significant after the first 110 min in any of the conditions; however, MTF duration was significantly higher than baseline in the standing conditions. Conclusion: Similar long-lasting weakening of MTF was induced by standing on a hard floor and on an antifatigue mat. However, walking partially attenuated this phenomenon. Application: Mostly static standing is likely to contribute to alterations of MTF in lower-leg muscles and potentially to musculoskeletal disorders regardless of the flooring characteristics. Occupational activities including slow-pace walking may reduce such deterioration in muscle function.

Sign in / Sign up

Export Citation Format

Share Document