# Wireless Body Area Network (WBAN)-Based Telemedicine for Emergency Care

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Related Work

## 3. Bayes Network Prediction for Telemedicine System

## 4. Bayes Network Prediction Model for Telemedicine

^{16}= 65,536 probabilities. However, through the CPT, there are only 28 conditional probabilities.

#### 4.1. Probabilities of Nodes (1 to 3)

#### 4.2. Probabilities of Nodes (4 to 14)

## 5. Results and Discussion

#### Manual Calculation

- P(A/B) = P(B/A) × P(B);
- As assumed through CPT, posterior probability P(IT|EM) = (P(EM|IT) × P(IT))/P(EM);
- (P(EM|IT) = 0.98 and P(IT) = 0.999;
- P(IT|EM) = (0.98 × 0.999)/P(EM);
- P(B) can be found using P(B) = P(B|A) × P(A) + P(B|~A) × P(~A);
- P(EM) = P(EM|IT) × P(IT) + P(EM|~IT) × P(~IT);
- P(EM) = 0.98 × 0.999 + 0.02 × 0.001 = 0.97904;
- Finally, the posterior probability is found using P(IT|EM) = (0.98 × 0.999)/0.97904 = 1.000.

- P(B|A) = sensitivity, true positive rate (TPR) = TP/(TP + FN) = P(EM|IT) = 0.98;
- P(B|~A) = false positive rate (FPR) = FP/(FP + TN) = P(EM|~IT) = 0.02;
- P(~B|~A) = specificity, true negative rate (TNR) = TN/(TN + FP) = P(~EM|~IT) = 0.96;
- P(~B|A) = false negative rate (FNR) = FN/(FN + TP) = P(~EM|IT = T) = 0.04.

- P(A) = probability of a positive class (PC) = P(IT) = 0.999;
- P(~A) = probability of a negative class (NC) = P(~IT) = 0.001;
- P(B) = probability of a positive prediction (PP) = P(EM) = 0.97904;
- P(~B) = probability of a negative prediction (NP) = P(~EM) = 0.02096.

- P(A|B) = (TPR × PC)/PP = (P(EM|IT) × P(IT))/P(EM);
- P(B) = TPR × PC + FPR × NC = P(EM|IT) × P(IT) + P(EM|~IT) × P(~IT).

- PPV = TP/(TP + FP);
- P(A|B) = PPV = TPR × PC/PP.

- P(IT|EM) = P(EM|IT) × P(IT)/P(EM);
- P(IT|EM) = 0.98 × 0.999/0.97904 = 0.999979.

## 6. Conclusions

## Author Contributions

## Funding

## Conflicts of Interest

## Appendix A

## References

- Misra, S.; Sarkar, S. Priority-based time-slot allocation in wireless body area networks during medical emergency situations: An evolutionary game-theoretic perspective. IEEE J. Biomed. Health Inform.
**2014**, 19, 541–548. [Google Scholar] [CrossRef] [PubMed] - Ross, P.E. Managing care through the air [remote health monitoring]. IEEE Spectr.
**2004**, 41, 26–31. [Google Scholar] [CrossRef] - Salayma, M.; Al-Dubai, A.; Romdhani, I.; Nasser, Y. Wireless body area network (WBAN) a survey on reliability, fault tolerance, and technologies coexistence. ACM Comput. Surv. (CSUR)
**2017**, 50, 1–38. [Google Scholar] [CrossRef][Green Version] - Bradai, N.; Chaari, L.; Kamoun, L. A comprehensive overview of wireless body area networks (WBAn). In Digital Advances in Medicine, E-Health, and Communication Technologies; IGI Global: Hershey, PA, USA, 2013; pp. 1–32. [Google Scholar]
- Elhadj, H.B.; Chaari, L.; Kamoun, L. A survey of routing protocols in wireless body area networks for healthcare applications. Int. J. E Health Med. Commun. (IJEHMC)
**2012**, 3, 1–18. [Google Scholar] [CrossRef] - Coping with the challenges of Designing Medical Body Area Networks. Available online: https://zmt.swiss/applications/wireless-body-area-networks/ (accessed on 14 January 2020).
- Meharouech, A.; Elias, J.; Mehaoua, A. Moving towards body-to-body sensor networks for ubiquitous applications: A survey. J. Sens. Actuator Netw.
**2019**, 8, 27. [Google Scholar] [CrossRef][Green Version] - Khan, R.A.; Pathan, A.S.K. The state-of-the-art wireless body area sensor networks: A survey. Int. J. Distrib. Sens. Netw.
**2018**, 14. [Google Scholar] [CrossRef] - Fallahpour, M. Wireless body area networking: Joint physical-networking layer simulation and modeling. In Medical Internet of Things (m-IoT)-Enabling Technologies and Emerging Applications; IntechOpen: London, UK, 2019. [Google Scholar]
- Cicioğlu, M.; Çalhan, A. SDN-based wireless body area network routing algorithm for healthcare architecture. ETRI J.
**2019**, 41, 452–464. [Google Scholar] [CrossRef][Green Version] - Maheswar, R.; Kanagachidambaresan, G.R.; Jayaparvathy, R.; Thampi, S.M. (Eds.) Body Area Network Challenges and Solutions; Springer: Berlin/Heidelberg, Germany, 2019. [Google Scholar]
- Evaluation of Wireless Body Area Networks. Available online: https://www.ijitee.org/wp-content/uploads/papers/v8i9S/I10560789S19.pdf (accessed on 20 January 2020).
- Negra, R.; Jemili, I.; Belghith, A. Wireless body area networks: Applications and technologies. Procedia Comput. Sci.
**2016**, 83, 1274–1281. [Google Scholar] [CrossRef][Green Version] - Nabi, M.; Geilen, M.; Basten, T. Wireless body area network data delivery. In Telemedicine and Electronic Medicine; CRC Press: Boca Raton, FL, USA, 2018; pp. 246–265. [Google Scholar]
- Bu, G.; Potop-Butucaru, M. Ban-gzkp: Optimal zero knowledge proof based scheme for wireless body area networks. Ad Hoc Netw.
**2018**, 77, 28–41. [Google Scholar] [CrossRef][Green Version] - Zuhra, F.T.; Bakar, K.A.; Ahmed, A.; Tunio, M.A. Routing protocols in wireless body sensor networks: A comprehensive survey. J. Netw. Comput. Appl.
**2017**, 99, 73–97. [Google Scholar] [CrossRef] - Shen, J.; Gui, Z.; Ji, S.; Shen, J.; Tan, H.; Tang, Y. Cloud-aided lightweight certificateless authentication protocol with anonymity for wireless body area networks. J. Netw. Comput. Appl.
**2018**, 106, 117–123. [Google Scholar] [CrossRef] - Hasan, K.; Biswas, K.; Ahmed, K.; Nafi, N.S.; Islam, M.S. A comprehensive review of wireless body area network. J. Netw. Comput. Appl.
**2019**, 143, 178–198. [Google Scholar] [CrossRef] - Samal, T.K.; Patra, S.C.; Kabat, M.R. An adaptive cuckoo search based algorithm for placement of relay nodes in wireless body area networks. J. King Saud Univ. Comput. Inf. Sci.
**2019**. [Google Scholar] [CrossRef] - Yi, C.; Zhao, Z.; Cai, J.; de Faria, R.L.; Zhang, G.M. Priority-aware pricing-based capacity sharing scheme for beyond-wireless body area networks. Comput. Netw.
**2016**, 98, 29–43. [Google Scholar] [CrossRef] - Liu, X.; Zhang, R.; Zhao, M. A robust authentication scheme with dynamic password for wireless body area networks. Comput. Netw.
**2019**, 161, 220–234. [Google Scholar] [CrossRef] - Arfaoui, A.; Kribeche, A.; Senouci, S.M.; Hamdi, M. Game-based adaptive anomaly detection in wireless body area networks. Comput. Netw.
**2019**, 163, 106870. [Google Scholar] [CrossRef] - Zhou, Y.; Sheng, Z.; Mahapatra, C.; Leung, V.C.; Servati, P. Topology design and cross-layer optimization for wireless body sensor networks. Ad Hoc Netw.
**2017**, 59, 48–62. [Google Scholar] [CrossRef][Green Version] - Kaur, N.; Singh, S. Optimized cost effective and energy efficient routing protocol for wireless body area networks. Ad Hoc Netw.
**2017**, 61, 65–84. [Google Scholar] [CrossRef] - Badreddine, W.; Khernane, N.; Potop-Butucaru, M.; Chaudet, C. Convergecast in wireless body area networks. Ad Hoc Netw.
**2017**, 66, 40–51. [Google Scholar] [CrossRef] - Zebboudj, S.; Cherifi, F.; Mohammedi, M.; Omar, M. Secure and efficient ECG-based authentication scheme for medical body area sensor networks. Smart Health
**2017**, 3, 75–84. [Google Scholar] [CrossRef] - Wang, W.; Shi, X.; Qin, T. Encryption-free authentication and integrity protection in body area networks through physical unclonable functions. Smart Health
**2019**, 12, 66–81. [Google Scholar] [CrossRef] - Mahmud, M.S.; Fang, H.; Carreiro, S.; Wang, H.; Boyer, E.W. Wearables technology for drug abuse detection: A survey of recent advancement. Smart Health
**2019**, 13, 100062. [Google Scholar] [CrossRef] - Silvera-Tawil, D.; Hussain, M.S.; Li, J. Emerging technologies for precision health: An insight into sensing technologies for health and wellbeing. Smart Health
**2019**, 15. [Google Scholar] [CrossRef] - Özderya, H.Y.; Erdöl, H.; Kayıkçıoğlu, T.; Yılmaz, A.Ö.; Kaya, İ. Wireless body area network studies for telemedicine applications using IEEE 802.15. 6 standard. In CMBEBIH; Springer: Singapore, Singapore, 2017; pp. 666–670. [Google Scholar]
- Agrahari, R.; Foroushani, A.; Docking, T.R.; Chang, L.; Duns, G.; Hudoba, M.; Karsan, A.; Zare, H. Applications of Bayesian network models in predicting types of hematological malignancies. Sci. Rep.
**2018**, 8, 6951. [Google Scholar] [CrossRef] [PubMed] - Bellot, D.; Boyer, A.; Charpillet, F.E.F. Designing smart agent based telemedicine systems using dynamic bayesian networks: An application to kidney disease people. In Proceedings of the 4th International Workshop on Enterprise Networking and Computing in Health Care Industry, Nancy, France, 6–7 June 2002; p. 8. [Google Scholar]
- Beretta, S.; Castelli, M.; Gonçalves, I.; Henriques, R.; Ramazzotti, D. Learning the structure of Bayesian Networks: A quantitative assessment of the effect of different algorithmic schemes. Complexity
**2018**, 1–12. [Google Scholar] [CrossRef] - Chan, F.; Wong, C.; Hon, T.; Choi, A. Bayesian network model for reducing accident rates of electrical and mechanical (E&M) work. Int. J. Environ. Res. Public Health
**2018**, 15, 2496. [Google Scholar] - Christophersen, A.; Deligne, N.I.; Hanea, A.M.; Chardot, L.; Fournier, N.; Aspinall, W.P. Bayesian network modeling and expert elicitation for probabilistic eruption forecasting: Pilot study for Whakaari/White Island, New Zealand. Front. Earth Sci.
**2018**, 6, 211. [Google Scholar] [CrossRef][Green Version] - Li, M.; Liu, K. Application of intelligent dynamic bayesian network with wavelet analysis for probabilistic prediction of storm track intensity index. Atmosphere
**2018**, 9, 224. [Google Scholar] [CrossRef][Green Version] - Pandey, S.K.; Mishra, R.B.; Triphathi, A.K. Software bug prediction prototype using bayesian network classifier: A comprehensive model. Procedia Comput. Sci.
**2018**, 132, 1412–1421. [Google Scholar] [CrossRef] - Park, E.; Chang, H.J.; Nam, H.S. A bayesian network model for predicting post-stroke outcomes with available risk factors. Front. Neurol.
**2018**, 9, 699. [Google Scholar] [CrossRef] - Scanagatta, M.; Corani, G.; De Campos, C.P.; Zaffalon, M. Approximate structure learning for large Bayesian networks. Mach. Learn.
**2018**, 107, 1209–1227. [Google Scholar] [CrossRef][Green Version] - Scutari, M.; Auconi, P.; Caldarelli, G.; Franchi, L. Bayesian networks analysis of malocclusion data. Sci. Rep.
**2017**, 7, 15236. [Google Scholar] [CrossRef] [PubMed][Green Version] - Wang, Y.; Cao, J.; Liu, L.; Feng, K.; Hong, S.; Xi, B.E.F. Framework of telemedicine diagnosis decision-making with Bayesian network based on multi-agent system. In Proceedings of the 7th International Conference on Computer Science & Education (ICCSE), Melbourne, VIC, Australia, 14–17 July 2012; pp. 68–70. [Google Scholar]
- Kammar, A.; Hernández-Hernández, M.; López-Moreno, P.; Ortíz-Bueno, A.; Martínez-Montaño, M. Probability and body composition of metabolic syndrome in young adults: Use of the bayes theorem as diagnostic evidence of the waist-to-height ratio. Stats
**2018**, 1, 3. [Google Scholar] [CrossRef][Green Version] - A Gentle Introduction to Bayes Theorem for Machine Learning. Available online: https://machinelearningmastery.com/bayes-theorem-for-machine-learning/ (accessed on 29 February 2020).
- A Gentle Introduction to Bayesian Belief Networks. Available online: https://machinelearningmastery.com/introduction-to-bayesian-belief-networks/ (accessed on 29 February 2020).

**Table 1.**Conditional probability table (CPT) for each node in the directed acyclic graph (DAG). IT—immediate teleconsultation; DS—delay-sensitive monitoring; GM—general monitoring; EM—emergency monitoring; VLL—very low latency; HC—high cost; LPLR—low packet loss rate; LL—low latency; VLPLR—very low packet loss rate; MDR—moderate data rate; HPLR—high packet loss rate; ML—moderate latency; MJ—moderate jitter; LJ—low jitter; HDR—high data rate; LC—low cost.

Node No. | Conditions | Probabilities |
---|---|---|

Node 1 | If IT is true | P(IT) = 0.999 |

Node 2 | If DS is true | P(DS) = 0.888 |

Node 3 | If GM is true | P(GM) = 0.777 |

Node 4 | If IT is true | P(EM) = 0.98 |

If IT is false | P(~EM) = 0.02 | |

Node 5 | If EM is true | P(VLL) = 0.90 |

If EM is false | P(~VLL) = 0.05 | |

Node 6 | If EM is true | P(HC) = 0.70 |

If EM is false | P(~HC) = 0.30 | |

Node 7 | If EM is true | P(LPLR) = 0.95 |

If EM is false | P(~LPLR) = 0.05 | |

Node 8 | If DS is true | P(LL) = 0.94 |

If DS is false | P(~LL) = 0.06 | |

Node 9 | If DS is true | P(VLPLR) = 0.97 |

If DS is false | P(~VLPLR) = 0.03 | |

Node 10 | If GM is true | P(MDR) = 0.99 |

If GM is false | P(~MDR) = 0.01 | |

Node 11 | If GM is true | P(HPLR) = 0.95 |

If GM is false | P(~HPLR) = 0.05 | |

Node 12 | If GM is true | P(ML) = 0.96 |

If GM is false | P(~ML) = 0.04 | |

Node 13 | If GM is true | P(MJ) = 0.94 |

If GM is false | P(~MJ) = 0.06 | |

Node 14 | If EM and DS are true | P(LJ) = 0.95; P(~LJ) = 0.05 |

If EM is true and DS is false | P(LJ) = 0.94; P(~LJ) = 0.06 | |

If EM is false and DS is true | P(LJ) = 0.74; P(~LJ) = 0.26 | |

If EM and DS are false | P(LJ) = 0.60; P(LJ) = 0.40 | |

Node 15 | If EM and DS are true | P(HDR) = 0.96; P(~HDR) = 0.04 |

If EM is true and DS is false | P(HDR) = 0.92; P(~HDR) = 0.08 | |

If EM is false and DS is true | P(HDR) = 0.74; P(~HDR) = 0.26 | |

If EM and DS are false | P(HDR) = 0.70; P(~HDR) = 0.30 | |

Node 16 | If DS and GM are true | P(LC) = 0.98; P(~LC) = 0.02 |

If DS is true and GM is false | P(LC) = 0.94; P(~LC) = 0.06 | |

If DS is false and GM is true | P(LC) = 0.70; P(~LC) = 0.30 | |

If DS and GM are false | P(LC) = 0.60; P(~LC) = 0.40 |

Positive Class | |
---|---|

Positive prediction | True positive (TP) |

Negative prediction | False negative (FN) |

Node No. | Conditions | Posterior Probabilities |
---|---|---|

Node 4 | P(EM) =$0.97904$ P(~EM) =$0.04092$ | P(IT|EM) = $0.999979$ P(IT|~EM) = $0.976539$ P(~IT|EM) = $0.00002$ P(~IT|~EM) = $0.02346$ |

Node 5 | P(VLL) =$0.883182$ P(~VLL) =$0.136778$ | P(EM|VLL) = $0.99768$ P(EM|~VLL) = $0.71578$ P(~EM|VLL) = $0.002316$ P(~EM|~VLL) = $0.284212$ |

Node 6 | P(HC) =$0.695118$ P(~HC) = $0.334222$ | P(EM|HC) =$0.98591$ P(EM|~HC) = $0.87879$ P(~EM|HC) = $0.000588$ P(~EM|~HC) = $0.121209$ |

Node 7 | P(LPLR) = 0.93172 P(~LPLR) = 0.088235 | P(EM| LPLR) = 0.998248 P(EM|~ LPLR) = 0.554791 P(~EM| LPLR) = 0.0017567 P(~EM|~ LPLR) = 0.44521 |

Node 8 | P(LL) = 0.84144 P(~LL) = 0.17408 | P(DS|LL) = 0.992013 P(DS|~LL) = 0.408088 P(~DS|LL) = 0.007986 P(~DS|~LL) = 0.591911 |

Node 9 | P(VLPLR) = 0.86472 P(~VLPLR) = 0.16632 | P(DS|VLPLR) = 0.98938 P(DS|~VLPLR) = 0.37121 P(~DS|VLPLR) = 0.003885 P(~DS|~VLPLR) = 0.62626 |

Node 10 | P(MDR) = 0.77146 P(~MDR) = 0.23408 | P(GM|MDR) = 0.997109 P(GM|~MDR) = 0.066387 P(~GM|MDR) = 0.0028906 P(~GM|~MDR) = 0.933612 |

Node 11 | P(HPLR) = 0.7493 P(~HPLR) = 0.23962 | P(GM|HPLR) = 0.98511 P(GM|~HPLR) = 0.09727 P(~GM|HPLR) = 0.014880 P(~GM|~HPLR) = 0.90272 |

Node 12 | P(ML) = 0.75454 P(~ML) = 0.22854 | P(GM|ML) = 0.98857 P(GM|~ML) = 0.033998 P(~GM|ML) = 0.01182 P(~GM|~ML) = 0.96600 |

Node 13 | P(MJ) = 0.74376 P(~MJ) = 0.2507 | P(GM|MJ) = 0.98201 P(GM|~MJ) = 0.15496 P(~GM|MJ) = 0.017989 P(~GM|~MJ) = 0.84503 |

Node No. | Conditions | Posterior Probabilities |
---|---|---|

Node 14 | If EM is true and DS is true and {P(LJ) = 0.95 and P(~LJ) = 0.05} If EM is true and DS is false and {P(LJ) = 0.94 and P(~LJ) = 0.06} If EM is false and DS is true and {P(LJ) = 0.74 and P(~LJ) = 0.26} If EM is false and DS is false and {P(LJ) = 0.60 and P(~LJ) = 0.40} | P(EM, DS|LJ) = 0.380 and P(EM, DS|~LJ) = 0.163 P(EM, ~DS|LJ) = 0.130 and P(EM, ~DS|~LJ) = 0.087 P(~EM, DS|LJ) = 0.108 and P(~EM, DS|~LJ) = 0.108 P(~EM, ~DS|LJ) = 0.087and P(~EM, ~DS|~LJ) = 0.222 |

Node 15 | If EM is true and DS is true and {P(HDR) = 0.96 and P(~HDR) = 0.04} If EM is true and DS is false and {P(HDR) = 0.92 and P(~HDR) = 0.08} If EM is true and DS is true and {P(HDR) = 0.74 and P(~HDR) = 0.26} If EM is true and DS is false and {P(HDR) = 0.70 and P(~HDR) = 0.30} | P(EM, DS|HDR) = 0.525 and P(EM, DS|~HDR) = 0.35 P(EM, ~DS|HDR) = 0.0625 and P(EM, ~DS|~HDR) = 0.0625 P(~EM, DS|HDR) = 0.0875 and P(~EM, DS|~HDR) = 0.0375 P(~EM, ~DS|HDR) = 0.1 and P(~EM, ~DS|~HDR) = 0.025 |

Node 16 | If DS is true and GM is true and {P(LC) = 0.98 and P(~LC) = 0.02} If DS is true and GM is false and {P(LC) = 0.94 and P(~LC) = 0.06} If DS is true and GM is true and {P(LC) = 0.70 and P(~LC) = 0.30} If EM is true and GM is false and {P(LC) = 0.60 and P(~LC) = 0.40} | P(DS, GM|LC) = 0.42 and P(DS, GM|~LC) = 0.18 P(DS, ~GM|LC) = 0.12 and P(DS, ~GM|~LC) = 0.08 P(~DS, GM|LC) = 0.05 and P(~DS, GM|~LC) = 0.05 P(~DS, ~GM|LC) = 0.08 and P(~DS, ~GM|~LC) = 0.02 |

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

R, L.; P, V.
Wireless Body Area Network (WBAN)-Based Telemedicine for Emergency Care. *Sensors* **2020**, *20*, 2153.
https://doi.org/10.3390/s20072153

**AMA Style**

R L, P V.
Wireless Body Area Network (WBAN)-Based Telemedicine for Emergency Care. *Sensors*. 2020; 20(7):2153.
https://doi.org/10.3390/s20072153

**Chicago/Turabian Style**

R, Latha, and Vetrivelan P.
2020. "Wireless Body Area Network (WBAN)-Based Telemedicine for Emergency Care" *Sensors* 20, no. 7: 2153.
https://doi.org/10.3390/s20072153