Automatic Diagnosis of Alzheimer’s Disease Using Neural Network Language Models

Fritsch, Julian; Wankerl, Sebastian; Nöth, Elmar

doi:10.1109/icassp.2019.8682690

Cited by 48 publications

(57 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Convo-lutional Neural Network (CNN) and their combination model were applied to extract linguistic features from transcripts of a picture description task [13]. In [14], two n-gram neural network language models for healthy controls (HC) and patients with dementia were built respectively for detection. It has been proven that, for the same database, neural networks achieved a better result compared with traditional machine learning methods [11,14].…”

Section: Related Workmentioning

confidence: 99%

“…In [14], two n-gram neural network language models for healthy controls (HC) and patients with dementia were built respectively for detection. It has been proven that, for the same database, neural networks achieved a better result compared with traditional machine learning methods [11,14]. The reason is, compared with deep learning methods, statistical models with handcrafted features are less robust and harder to design, as it requires more expert knowledge.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Hierarchical Attention Neural Network for Detecting AD

Pan¹,

Mirheidari²,

Reuber³

et al. 2019

Interspeech 2019

View full text Add to dashboard Cite

Picture description tasks are used for the detection of cognitive decline associated with Alzheimer's disease (AD). Recent years have seen work on automatic AD detection in picture descriptions based on acoustic and word-based analysis of the speech. These methods have shown some success but lack an ability to capture any higher level effects of cognitive decline on the patient's language. In this paper, we propose a novel model that encompasses both the hierarchical and sequential structure of the description and detect its informative units by attention mechanism. Automatic speech recognition (ASR) and punctuation restoration are used to transcribe and segment the data. Using the DementiaBank database of people with AD as well as healthy controls (HC), we obtain an F-score of 84.43% and 74.37% when using manual and automatic transcripts respectively. We further explore the effect of adding additional data (a total of 33 descriptions collected using a 'digital doctor') during model training, and increase the F-score when using ASR transcripts to 76.09%. This outperforms baseline models, including bidirectional LSTM and bidirectional hierarchical neural network without an attention mechanism, and demonstrate that the use of hierarchical models with attention mechanism improves the AD/HC discrimination performance.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Automatic Hierarchical Attention Neural Network for Detecting AD

Pan¹,

Mirheidari²,

Reuber³

et al. 2019

Interspeech 2019

View full text Add to dashboard Cite

show abstract

“…At the same time, this model achieved the best precision and F 1 scores with 6.55% and 2.80% improvements, respectively. Still, Fritsch et al [30] showed the best recall score with 1.66% difference although they did not report F 1 measure . The first advantage of our proposed methods compared to Fritsch et al [30] is that we train a single language model for both the AD and HC groups which helps the model to use samples from both classes for the desired task.…”

Section: Interpretation Of Resultsmentioning

confidence: 92%

“…Still, Fritsch et al [30] showed the best recall score with 1.66% difference although they did not report F 1 measure . The first advantage of our proposed methods compared to Fritsch et al [30] is that we train a single language model for both the AD and HC groups which helps the model to use samples from both classes for the desired task. The other advantage is that our models are highly pre-trained on large datasets which enables them to start training on new, smaller datasets with good initialization parameters and also avoid overfitting.…”

Section: Interpretation Of Resultsmentioning

confidence: 92%

Transformer-based deep neural network language models for Alzheimer’s disease detection from targeted speech

Roshanzamir

Aghajan

Baghshah

2020

Preprint

View full text Add to dashboard Cite

Background: We developed transformer-based deep learning models based on natural language processing for early diagnosis of Alzheimer’s disease from the picture description test.Methods: The lack of large datasets poses the most important limitation for using complex models that do not require feature engineering. Transformer-based pre-trained deep language models have recently made a large leap in NLP research and application. These models are pre-trained on available large datasets to understand natural language texts appropriately, and are shown to subsequently perform well on classification tasks with small training sets. The overall classification model is a simple classifier on top of the pre-trained deep language model.Results: The models are evaluated on picture description test transcripts of the Pitt corpus, which contains data of 170 AD patients with 257 interviews and 99 healthy controls with 243 interviews. The large bidirectional encoder representations from transformers (BERTLarge) embedding with logistic regression classifier achieves classification accuracy of 88.08%, which improves thestate-of-the-art by 2.48%.Conclusions: Using pre-trained language models can improve AD prediction. This not only solves the problem of lack of sufficiently large datasets, but also reduces the need for expert-defined features.

show abstract

“…Fritsch et al [30] used two different auto-regressive LSTM-based neural network language models to classify AD and HC transcripts of the Pitt corpus from the DementiaBank dataset. After that, Pan et al [31] worked on predicting AD using a stacked bidirectional LSTM and gated recurrent unit (GRU) layers equipped with a hierarchical attention mechanism.…”

Section: Deep Learning-based Approachesmentioning

confidence: 99%

Transformer-based deep neural network language models for Alzheimer's disease risk assessment from targeted speech

Roshanzamir

Aghajan

Baghshah

2021

Preprint

View full text Add to dashboard Cite

Background: We developed transformer-based deep learning models based on natural language processing for early risk assessment of Alzheimer’s disease from the picture description test.Methods: The lack of large datasets poses the most important limitation for using complex models that do not require feature engineering. Transformer-based pre-trained deep language models have recently made a large leap in NLP research and application. These models are pre-trained on available large datasets to understand natural language texts appropriately, and are shown to subsequently perform well on classiﬁcation tasks with small training sets. The overall classiﬁcation model is a simple classiﬁer on top of the pre-trained deep language model.Results: The models are evaluated on picture description test transcripts of the Pitt corpus, which contains data of 170 AD patients with 257 interviews and 99 healthy controls with 243 interviews. The large bidirectional encoder representations from transformers (BERTLarge) embedding with logistic regression classiﬁer achieves classiﬁcation accuracy of 88.08%, which improves the state-of-the-art by 2.48%.Conclusions: Using pre-trained language models can improve AD prediction. This not only solves the problem of lack of suﬃciently large datasets, but also reduces the need for expert-deﬁned features.

show abstract

Automatic Diagnosis of Alzheimer’s Disease Using Neural Network Language Models

Cited by 48 publications

References 10 publications

Automatic Hierarchical Attention Neural Network for Detecting AD

Automatic Hierarchical Attention Neural Network for Detecting AD

Transformer-based deep neural network language models for Alzheimer’s disease detection from targeted speech

Transformer-based deep neural network language models for Alzheimer's disease risk assessment from targeted speech

Contact Info

Product

Resources

About