Adaptation of IDPT System Based on Patient-Authored Text Data using NLP

Mukhiya, Suresh Kumar; Ahmed, Usman; Rabbi, Fazle; Pun, Ka I; Lamo, Yngve

doi:10.1109/cbms49503.2020.00050

Cited by 20 publications

(16 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 1 seed terms are extended using the Wordnet method. There are various lists of depression symptoms in different classification systems (Mukhiya et al, 2020a ). These lists use either clinical or casual symptoms terms depending on whether the poll is a questionnaire to the patient or the clinician.…”

Section: Methodsmentioning

confidence: 99%

“…These lists use either clinical or casual symptoms terms depending on whether the poll is a questionnaire to the patient or the clinician. Major classification systems for depression such as DSM-V 4 and ICD-10 (World Health Organization, 1993 ) are widely used depression scales that were merged to deduce a fine base list of symptoms (Mukhiya et al, 2020a ).…”

Section: Methodsmentioning

confidence: 99%

“…The remaining data are annotated by using the proposed embedding method. The labeling is done using the PHQ-9 rating method, i.e., such that 0 indicates not depressed, 1 mildly depressed, 2 moderately depressed, and 3 severely depressed (Mukhiya et al, 2020a ). We convert the annotation into a binary class for each symptom, where 0 indicates the absence of symptoms and 1 indicates the presence of symptoms.…”

Section: Methodsmentioning

confidence: 99%

“…The dataset is gathered from an online forum, website, and social media site (Mukhiya et al, 2020a). Amazon Mechanical Turk 5 service is used to label the 500 texts (Mukhiya et al, 2020a). The remaining data are annotated by using the proposed embedding 5 https://www.mturk.com/.…”

Section: Datasetmentioning

confidence: 99%

“…Extracting useful knowledge is still difficult because of these conflicting reports (Ebadi et al, 2020 ). A combination of recent events and longer-term and/or personal factors trigger depression rather than just a single immediate issue or event (Mukhiya et al, 2020a ). Although everyone is different, identifying the cause or change in difficult circumstances cannot be possible always (Losada and Gamallo, 2018 ).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Attention-Based Deep Entropy Active Learning Using Lexical Algorithm for Mental Health Treatment

et al. 2021

Self Cite

View full text Add to dashboard Cite

With the increasing prevalence of Internet usage, Internet-Delivered Psychological Treatment (IDPT) has become a valuable tool to develop improved treatments of mental disorders. IDPT becomes complicated and labor intensive because of overlapping emotion in mental health. To create a usable learning application for IDPT requires diverse labeled datasets containing an adequate set of linguistic properties to extract word representations and segmentations of emotions. In medical applications, it is challenging to successfully refine such datasets since emotion-aware labeling is time consuming. Other known issues include vocabulary sizes per class, data source, method of creation, and baseline for the human performance level. This paper focuses on the application of personalized mental health interventions using Natural Language Processing (NLP) and attention-based in-depth entropy active learning. The objective of this research is to increase the trainable instances using a semantic clustering mechanism. For this purpose, we propose a method based on synonym expansion by semantic vectors. Semantic vectors based on semantic information derived from the context in which it appears are clustered. The resulting similarity metrics help to select the subset of unlabeled text by using semantic information. The proposed method separates unlabeled text and includes it in the next active learning mechanism cycle. Our method updates model training by using the new training points. The cycle continues until it reaches an optimal solution, and it converts all the unlabeled text into the training set. Our in-depth experimental results show that the synonym expansion semantic vectors help enhance training accuracy while not harming the results. The bidirectional Long Short-Term Memory (LSTM) architecture with an attention mechanism achieved 0.85 Receiver Operating Characteristic (ROC curve) on the blind test set. The learned embedding is then used to visualize the activated word's contribution to each symptom and find the psychiatrist's qualitative agreement. Our method improves the detection rate of depression symptoms from online forum text using the unlabeled forum texts.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Datasetmentioning

confidence: 99%