Leveraging Deep Learning for Designing Healthcare Analytics Heuristic for Diagnostics

Shafqat, Sarah; Fayyaz, Maryyam; Khattak, Hasan Ali; Bilal, Muhammad; Khan, Shariq Mahmood; Ishtiaq, Osama; Abbasi, Almas; Shafqat, Farzana; Alnumay, Waleed S.; Chatterjee, Pushpita

doi:10.1007/s11063-021-10425-w

Cited by 18 publications

(7 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dengue critical phase, which lasts 24–48 hours and begins with defervescence, is the most dangerous phase of the disease. Severe dengue symptoms, such as plasma leakage or fluid accumulation, as well as respiratory problems, serious bleeding, or organ malfunction, can result in death [ 9 ]. A Chikungunya infection can induce symptoms like those of dengue fever, but with more severe joint pain and swelling.…”

Section: Introductionmentioning

confidence: 99%

Recurrent Neural Networks for Feature Extraction from Dengue Fever

Daniel¹,

Sherly²,

Ponnuramu

et al. 2022

Evidence-Based Complementary and Alternative Medicine

View full text Add to dashboard Cite

Dengue fever modelling in endemic locations is critical to reducing outbreaks and improving vector-borne illness control. Early projections of dengue are a crucial tool for disease control because of the unavailability of treatments and universal vaccination. Neural networks have made significant contributions to public health in a variety of ways. In this paper, we develop a deep learning modelling using random forest (RF) that helps extract the features of the dengue fever from the text datasets. The proposed modelling involves the data collection, preprocessing of the input texts, and feature extraction. The extracted features are studied to test how well the feature extraction using RF is effective on dengue datasets. The simulation result shows that the proposed method achieves higher degree of accuracy that offers an improvement of more than 12% than the existing methods in extracting the features from the input datasets than the other feature extraction methods. Further, the study reduces the errors associated with feature extraction that is 10% lesser than the other existing methods, and this shows the efficacy of the model.

show abstract

Section: Introductionmentioning

confidence: 99%

Recurrent Neural Networks for Feature Extraction from Dengue Fever

Daniel¹,

Sherly²,

Ponnuramu

et al. 2022

Evidence-Based Complementary and Alternative Medicine

View full text Add to dashboard Cite

show abstract

“…It was cleaned and pruned where the target diagnostic labels were missing. We started off with some traditional ML algorithms like; multinomial logistic regression, decision tree, naïve Bayes, ada boost and light gradient boosting machine (Light GBM) as in [40]. Our previous explorations in [40] and [41] showed us some good results using deep learning heuristics.…”

Section: Related Workmentioning

confidence: 99%

“…We started off with some traditional ML algorithms like; multinomial logistic regression, decision tree, naïve Bayes, ada boost and light gradient boosting machine (Light GBM) as in [40]. Our previous explorations in [40] and [41] showed us some good results using deep learning heuristics. ML algorithms integrated with traditional NLP methods were also experimented with and results were obtained.…”

Section: Related Workmentioning

confidence: 99%

NER Sequence Embedding of Unified Medical Corpora to Incorporate Semantic Intelligence in Big Data Healthcare Diagnostics

Shafqat,

Anwar,

Javaid

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Clinical diagnosis is a challenging task for which high expertise is required at the doctors’ end. It is recognized that technology integration with the clinical domain would facilitate the diagnostic process. A semantic understanding of the medical domain and clinical context is needed to make intelligent analytics. These analytics need to learn the medical context for different purposes of diagnosing and treating patients. Traditional diagnoses are made through phenotype features from patients’ profiles. It is also a known fact that diabetes mellitus (DM) is widely affecting the population and is a chronic disease that requires timely diagnosis. The motivation for this research comes from the gap found in discovering the common ground for medical context learning in analytics to diagnose DM and its comorbidity diseases. Therefore, a unified medical knowledge base is found significantly important to learning contextual Named Entity Recognition (NER) embedding for semantic intelligence. Researchers in this paper have searched for possible solutions for medical context learning and found that unified corpora tagged with medical terms were missing to train the analytics for diagnoses of DM and its comorbidities. Hence, effort was put into collecting endocrine diagnostic electronic health records (EHR) corpora for clinical purposes that are manually labeled with ICD-10-CM international coding scheme to minimise chances of error. International Codes for Diseases (ICD) by the World Health Organization (WHO) is a known schema to represent medical codes for diagnoses. The complete endocrine EHR corpora make DM-Comorbid-EHR-ICD-10 Corpora. DM-Comorbid-EHR-ICD-10 Corpora is tagged for understanding the medical context with uniformity. In this research experiments were run with different NER sequence embedding approaches using advanced ML integrated with NLP techniques. These experiments used common frameworks like; Spacy, Flair, and TensorFlow, Keras. These experiments led to yield albeit label sets in the form of (instance, label) pair for diagnoses that were tagged with the Sequential() model found in TensorFlow.Keras using Bi-LSTM and dense layers. The maximum accuracy achieved was 0.9 for Corpus14407_DM_pts_33185 with a maximum number of diagnostic features taken as input. The sequential DNN NER model diagnostic accuracy increased as the size of the corpus grew from 100 to 14407 DM patients suffering from comorbidity diseases. The significance of clinical notes and practitioner comments available as free text is clearly seen in the diagnostic accuracy.

show abstract

“…A specific resource library includes specific learning objects. The standard of data collection can be described concerning specific metadata, and the online learning resources encapsulated by it can be identified and effectively located through the network to achieve specific semantic labeling [15][16]. Some online learning resources are shown in Figure 3.…”

Section: Description Of Algorithm Experiments Processmentioning

confidence: 99%

Personalized Learning Resource Push Evaluation Considering Image Big Data Processing Technology

2024

IJCAI

View full text Add to dashboard Cite

With the continuous development of the social economy, the ways of obtaining learning materials and learning resources have gradually increased, especially the development of the Internet has enabled more high-quality resources to be shared, but it should be noted that it is precisely because of the excessive resources, it is relatively complicated to find suitable and interesting learning resources. In response to these needs and deficiencies, this paper introduce image big data processing technology, characterizes specific learning characteristics using the time decay function by sorting out the business logic of personalized learning resource push. It also imports specific learning cognitive levels, and matches with the corresponding learning resources, maximizes the quality service of personalized Learning facilities, and improves the efficiency of learning resources. The study findings demonstrate that the image big data processing technique is effective and can support the push evaluation of personalized learning resources.Povzetek: Prispevek ocenjuje personalizirano posredovanje učnih virov ob upoštevanju tehnologije obdelave slikovnih velikih podatkov. Z uporabo funkcije časovnega razpada za karakterizacijo učnih značilnosti in uvozom specifičnih učnih kognitivnih ravni se izboljša učinkovitost in kakovost posredovanja učnih virov. Študija potrjuje učinkovitost te tehnologije pri podpori ocenjevanja personaliziranih učnih virov.

show abstract

Leveraging Deep Learning for Designing Healthcare Analytics Heuristic for Diagnostics

Cited by 18 publications

References 53 publications

Recurrent Neural Networks for Feature Extraction from Dengue Fever

Recurrent Neural Networks for Feature Extraction from Dengue Fever

NER Sequence Embedding of Unified Medical Corpora to Incorporate Semantic Intelligence in Big Data Healthcare Diagnostics

Personalized Learning Resource Push Evaluation Considering Image Big Data Processing Technology

Contact Info

Product

Resources

About