ALSNet: A Dilated 1-D CNN for Identifying ALS from Raw EMG Signal

Hassan, Kanza; Hridoy, Md. Shamiul Alam; Tasnim, Naima; Chowdhury, Atia Faria; Roni, Tanvir Alam; Tabrez, Sheikh; Subhana, Arik; Shahnaz, Celia

doi:10.1109/icassp43922.2022.9747366

Cited by 4 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Raw EMG data are inherently complex and contains a vast amount of information, requiring substantial computational resources and time for analysis. 30 Principal component analysis (PCA) provides a way to simplify the complexity and variety in EMG data, while preserving trends and patterns. 31 The major goal of PCA is to reduce the high "dimensionality" of these data, without losing important information.…”

Section: Emg Signal Feature Selectionmentioning

confidence: 99%

“…DL models are capable of effectively handling the substantial amount of information present in raw EMG data, resulting in enhanced performance. 30,37 The subsequent sections will explore these advancements and highlight the superior capabilities of DL models. 39 This small dataset, despite its popularity [39][40][41] does not fulfill the growing need for representatively diverse and comprehensive datasets in the EDX field.…”

Section: Emg Signal Feature Selectionmentioning

confidence: 99%

“…However, recent advancements in DL have better facilitated the utilization of raw data as input for EMG analysis. DL models are capable of effectively handling the substantial amount of information present in raw EMG data, resulting in enhanced performance 30,37 . The subsequent sections will explore these advancements and highlight the superior capabilities of DL models.…”

Section: Ai In Edx Medicinementioning

confidence: 99%

“…The recent advancements allowing DL models to classify raw EMG signals without the need for manual feature extraction allow for faster and real‐time application, with overall less computing capacity needed. One notable example is ALSNet, 30 composed of three convolutional layers, which were trained using data collected from 10 healthy individuals and 8 patients with ALS. Remarkably, this model achieved an impressive accuracy of 97.7% in differentiating between 1‐s clips of normal and ALS patient EMG signals.…”

Section: Ai In Edx Medicinementioning

confidence: 99%

See 3 more Smart Citations

The role of artificial intelligence in electrodiagnostic and neuromuscular medicine: Current state and future directions

Taha,

Morren

2023

Muscle and Nerve

View full text Add to dashboard Cite

The rapid advancements in artificial intelligence (AI), including machine learning (ML), and deep learning (DL) have ushered in a new era of technological breakthroughs in healthcare. These technologies are revolutionizing the way we utilize medical data, enabling improved disease classification, more precise diagnoses, better treatment selection, therapeutic monitoring, and highly accurate prognostication. Different ML and DL models have been used to distinguish between electromyography signals in normal individuals and those with amyotrophic lateral sclerosis and myopathy, with accuracy ranging from 67% to 99.5%. DL models have also been successfully applied in neuromuscular ultrasound, with the use of segmentation techniques achieving diagnostic accuracy of at least 90% for nerve entrapment disorders, and 87% for inflammatory myopathies. Other successful AI applications include prediction of treatment response, and prognostication including prediction of intensive care unit admissions for patients with myasthenia gravis. Despite these remarkable strides, significant knowledge, attitude, and practice gaps persist, including within the field of electrodiagnostic and neuromuscular medicine. In this narrative review, we highlight the fundamental principles of AI and draw parallels with the intricacies of human brain networks. Specifically, we explore the immense potential that AI holds for applications in electrodiagnostic studies, neuromuscular ultrasound, and other aspects of neuromuscular medicine. While there are exciting possibilities for the future, it is essential to acknowledge and understand the limitations of AI and take proactive steps to mitigate these challenges. This collective endeavor holds immense potential for the advancement of healthcare through the strategic and responsible integration of AI technologies.

show abstract

Section: Emg Signal Feature Selectionmentioning

confidence: 99%