kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Classification of Pain level using Zygomaticus and Corrugator EMG Features: Machine Learning Approach
KTH, School of Electrical Engineering and Computer Science (EECS), Electrical Engineering, Electronics and Embedded systems, Integrated devices and circuits.ORCID iD: 0000-0003-2357-1108
KTH, School of Electrical Engineering and Computer Science (EECS), Electrical Engineering, Electronics and Embedded systems, Integrated devices and circuits.ORCID iD: 0000-0003-1959-6513
(English)In: Article in journal (Refereed) Submitted
Abstract [en]

A real-time recognition of facial expressions is required to certify the accurate pain assess-8 ment of patients in ICU, infants, and other patients who may not be able to communicate verbally 9 or even express the sensation of pain. Facial expression is a key pain-related behavior that may 10 unlock the answer to objective pain measurement tool. In this work, a machine learning based pain 11 level classification using data collected from facial electromyograms (EMG) is presented. The da-12 taset is acquired from part of Bio Vid Heat Pain database [1] to evaluated facial expression from emg 13 corrugator and emg zygomaticus and an EMG signal processing and data analysis flow is adapted 14 for continuous pain estimation. The extracted pain-associated facial electromyography (fEMG) fea-15 tures classification is performed by a supervised ML algorithm, on the KNN by choosing the value 16 of k and that depends on the nonlinear models. The presentation of the accuracy estimation is per-17 formed with and considerable growth in classification accuracy is noticed when the subject matter 18 from the features is omitted from the analysis. The ML algorithm for classification of the amount of 19 pain in patients could deliver valuable evidence for the health care providers and aid the treatment 20 assessment. Performances of 99.4% shown on the binary classification for the dis-crimination be-21 tween the baseline and the pain tolerance level (P0 verse P4) without the influence of on a subject 22 bias. Moreover, the result of the classification accuracy is clearly showing the relevance of the pro-23 posed approach.

Keywords [en]
facial electromyograms (fEMG); machine learning classification; pain intensity. 25
National Category
Engineering and Technology Electrical Engineering, Electronic Engineering, Information Engineering Signal Processing
Research subject
Electrical Engineering
Identifiers
URN: urn:nbn:se:kth:diva-304353OAI: oai:DiVA.org:kth-304353DiVA, id: diva2:1611285
Note

QCR 20211124

Available from: 2021-11-14 Created: 2021-11-14 Last updated: 2022-06-25Bibliographically approved
In thesis
1. Data-driven Implementations for Enhanced Healthcare Internet-of-Things Systems
Open this publication in new window or tab >>Data-driven Implementations for Enhanced Healthcare Internet-of-Things Systems
2021 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Healthcare monitoring systems based on the Internet of Things (IoT) areemerging as a potential solution for reducing healthcare costs by impacting and improving the quality of health care delivery. The rising numberof elderly and chronic patient population in the world and the associatedhealthcare costs urges the application of IoT technology to improve andsupport the health care services. This thesis develops and integrates twoIoT-based healthcare systems aiming to support elderly independent livingat home. The first one involves using IoT-based remote monitoring for paindetection, while the second one detects behavioral changes caused by illnessvia profiling the appliances’ energy usage.In the first approach, an Electromyography (EMG )sensor node with aWireless Fidelity (Wi-Fi) radio module is designed for monitoring the painof patients living at home. An appropriate feature-extraction and classification algorithm is applied to the EMG signal. The classification algorithmachieves 98.5% accuracy for the experimental data collected from the developed EMG sensor node, while it achieves 99.4% classification accuracy forthe clinically approved pain intensity dataset. Moreover, the experimentalresults clearly show the relevance of the proposed approaches and provetheir suitability for real-life applications. The developed sensor node for thepain level classification method is beneficial for continuous pain assessmentto the smart home-care community.As a complement to the first approach, in the second approach, an IoTbased smart meter and a set of appliance-level load profiling methods aredeveloped to detect the electricity usage of users’ daily living at home, whichindirectly provides information about the subject’s health status. The thesishas formulated a novel methodology by integrating Non-intrusive ApplianceLoad Monitoring (NIALM) analysis with Machine Learning- (ML) basedclassification at the fog layer. The developed method allows the detectionof a single appliance with high accuracy by associating the user’s Activitiesof Daily Living (ADL). The appliances detection is performed by employinga k-Nearest Neighbors (k-NN) classification algorithm. It achieves 97.4% accuracy, demonstrating its high detection performance. Due to the low cost and reusability advantages of Field Programmable Gate Arrays (FPGA),the execution of k-NN for appliances classification model is performed onan FPGA. Its classification performance was comparable with other computing platforms, making it a cost-effective alternative for IoT-based healthcare assessment of daily living at home. The developed methods have haspractical application in assisting real-time e-health monitoring of any individual who can remain in the comfort of their normal living environment. 

Abstract [sv]

System för monitorering av hälso- och sjukvård baserade på IoT (internet of things) erbjuder idag kostnadseffektiva lösningar som många gånger kan utgöra bättre alternativ än traditionell övervakning inom vård och omsorg. Kostnader för sjukvård stiger brant, mycket på grund av en ökande andel äldre i befolkningen, och kraven på sjukhus och vårdinstanser att tillhandahålla högkvalitativa tjänster stiger därmed och blir alltmer utmanande. Denna avhandling presenterar två olika integrerade IoT-system, som utvecklats för att monitorera hälsotillståndet hos vårdbehövande och äldre personer i deras hemmiljö. Det första systemet bygger på en fjärransluten IoT-lösning för smärta, medan det andra upptäcker förändrade levnadsmönster som orsakas av sjukdom genom att monitorera el-användningen för den vårdbehövande.

I den första varianten har en elektromyografi (EMG) sensor med en wifi-modul designats för att övervaka smärtkänningar hos hemmaboende patienter. En algoritm extraherar relevanta data ur EMG signalen och utvärderar dessa för att kunna ange den smärtnivå som patienten upplever. Denna process ger 98.5 % rätt angivna smärtnivåer hos den uppmätta signalen från EMG-sensorn, men hela 99.4 % rättbestämda smärtnivåer från det kliniskt godkända testet av smärtnivå Bio Vid. De experimentella resultaten visar tydligt att den föreslagna metoden lämpar sig utmärkt för fortsatta försök på människor.

I den andra IoT-lösningen, som ska ses som ett komplement till den första, används en IoT-baserad smart mätare tillsammans med en uppsättning metoder för att bestämma belastningsprofilen för elanvändningen i den vårdbehövandes bostad och därigenom indirekt upptäcka avvikelser som indikerar att hälsotillståndet hos den inneboende har förändrats. I avhandlingen har en ny metodologi införts kallad ”non-intrusive appliance load monitoring (NIALM), som baseras på maskininlärd klassificering av ”fog layer”. Metoden gör det möjligt att urskilja enskilda el-konsumenter med stor noggrannhet genom att jämföra mätdata med hushållets ”activities of daily living”, (ADL). Detektionen av olika el-konsumenter i hushållet görs genom klassificeringsalgoritmen ”k-nearest neighbour’s” (k-NN), vilken har uppnått hela 97.4% träffsäkerhet och tydligt demonstrerar metodens användbarhet. Tack vare de låga kostnaderna och möjligheten till åter-användning hos ”field programmable gate arrays” (FPGA), har den k-NN-baserade modellen implementerats i FPGA. Prestandan för detta system visar sig, vid jämförelse med andra beräkningsplattformar, vara ett kostnadseffektivt sätt att använda IoT-baserade lösningar för monitorering av personers hälsostatus i hemmiljö.

Sammanfattningsvis visar avhandlingen på två integrerade IoT-lösningar för patientövervakning i hemmiljö, som kombinerar smärtupplevelser med ADL och därigenom kan erbjuda trygg och kostnadseffektiv assistans till vården av sjuka och äldre personer och möjliggöra för individer att leva längre i sina egna hem.

Place, publisher, year, edition, pages
Stockholm, Sweden: KTH Royal Institute of Technology, 2021. p. 122
Series
TRITA-EECS-AVL
Keywords
IoT, Real_time, Smart meter, Biosignal, NIALM, Data-driven, Machine Learning, FPGA
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Information and Communication Technology
Identifiers
urn:nbn:se:kth:diva-304961 (URN)978-91-8040-008-4 (ISBN)
Public defence
2021-12-10, Sal C (Sal Sven-Olof Ohrvik) /zoom online link: https://kth-se.zoom.us/j/69579814475?pwd=L3ZNTmVjeWFBRWtZLythZTFnRmpOQT09, Kistagången 16, Kista., Stockholm, 10:00 (English)
Opponent
Supervisors
Note

QC 20211119

Available from: 2021-11-19 Created: 2021-11-19 Last updated: 2022-06-25Bibliographically approved

Open Access in DiVA

No full text in DiVA

Authority records

Kelati, AmlesetPlosila, JuhaTenhunen, Hannu

Search in DiVA

By author/editor
Kelati, AmlesetPlosila, JuhaTenhunen, Hannu
By organisation
Integrated devices and circuits
Engineering and TechnologyElectrical Engineering, Electronic Engineering, Information EngineeringSignal Processing

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 114 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf