Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM64863
Volume:
10 |
6 |
Jun 2026
Pages:
1-6
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Department of Computer Science Engineering
Galgotias University
Department of Computer Science Engineering
Galgotias University
Artificial intelligence (AI) in telemedicine is trans-forming the delivery of healthcare by automating patient intake and streamlining clinical workflows. In this paper, we analyze the importance of AI-driven chatbots in modern healthcare through architectural and empirical analysis of MedCore AI, a full-stack telemedicine ecosystem. The platform integrates a Google Gemini powered medical triage engine, secure doctor management and real-time Socket.IO chat messaging and WebRTC video calling. MedCore AI is built on a classic three-tier architecture with React on the frontend, Node.js/Express on the backend, and MongoDB for the database layer. In this context, the embedded AI chatbot becomes the center of engagement processing the raw patient symptom inputs, assigning the emergency triage classifications, embedding live search grounding links, and gen-erating structured summaries for verified physicians. Empirical performance metrics demonstrate great efficiency in internal data transactions. Average Indexed MongoDB single collection DB queries run at 20 ms . Real-time chat synchronization runs smoothly at 100 ms . Normal REST API requests run at 250 ms . However, the platform suffers from latency bottlenecks in complex asynchronous integration scenarios. The AI triage chatbot suffers from an average delay of 1800 ms (1.2-3.5 seconds) due to the computational cost of natural language generation and search grounding. Moreover, the WebRTC video consultation setup shows the highest latency spike of 4000 ms, which is a signature of the heavy network signaling and media negotiation in STUN/TURN configurations. In summary, this study highlights that direct embedding of automated conversational AI into a full-stack ecosystem significantly reduces clinical intake friction. The integration of intelligent automated triage with persistent communication channels offers a highly optimized and MVP-ready framework that overcomes the latency of external APIs and significantly improves the delivery of digital healthcare.
KEYWORDS
Putting together an online medical advice system means checking whether its main processes work properly. Instead of just guessing, there is a review showing how each part performs - proving key functions now run live. Right here fits Table I, called Feature Evaluation of MedCore AI, placed to back up claims with real results. Seeing it helps trust what the system can actually do. That table belongs in this spot, no elsewhere. Right off the bat, the setup includes every key part needed for smooth telehealth service. Instead of waiting, patients get a first check through smart software that guides them based on symptoms. Behind the scenes, doctors are confirmed securely, appointments line up across one clear schedule, payments move without hiccups - all working as intended. At the core, talking to care providers happens instantly, messages stick around, chats stay active, video links hold strong. What stands out is how each piece runs without gaps, fitting together like parts meant to be used daily. Putting TABLE I right here shows the system works in real life, not just on paper, handling full conversations between doctors and patients. Proof sits ready because this table confirms the foundation runs properly before diving into speed tests and how well it performs.
[1] A. Kapoor and S. Dey, “Conversational AI in Healthcare: A Review of Large Language Models for Patient Triage,” IEEE Transactions on Medical Informatics, vol. 42, no. 3, pp. 512–520, Mar. 2024. [2] P. Llanos-Sa´nchez, A. Espinoza-Franco, F. Portugue´z-Villa, G. Ferna´ndez-Gutie´rrez, and A. Pacheco, “Medibot AI: a web-based soft-ware tool integrating Gemini API for informational over-the-counter medication guidance,” F1000Research, vol. 15, no. 572, Apr. 2026. doi: 10.12688/f1000research.177717.1. [3] D. A. Dhinagaran, L. Martinengo, M.-H. R. Ho, S. Joty, T. Kowatsch,R. Atun, and L. T. Car, “Designing, Developing, Evaluating, and Im-plementing a Smartphone-Delivered, Rule-Based Conversational Agent (DISCOVER): Development of a Conceptual Framework,” JMIR mHealth and uHealth, vol. 10, p. e38740, Oct. 2022. doi: 10.2196/38740. [4] S. M. Freire, D. Teodoro, F. Wei-Kleiner, E. Sundvall, D. Karlsson, and P. Lambrix, “Comparing the Performance of NoSQL Approaches for Managing Archetype-Based Electronic Health Record Data,” PLOS ONE, vol. 11, no. 3, p. e0150069, Mar. 2016. doi: 10.1371/jour-nal.pone.0150069. [5] D. Skvorc, M. Horvat, and S. Srbljic, “Performance evaluation of Websocket protocol for implementation of full-duplex web streams,” in 2014 37th International Convention on Information and Communi-cation Technology, Electronics and Microelectronics (MIPRO), Opatija, Croatia, 2014, pp. 1003–1008. doi: 10.1109/mipro.2014.6859715. [6] A. Rahmatulloh, I. Darmawan, and R. Gunawan, “Performance Analysis of Data Transmission on WebSocket for Real-time Communication,” in 2019 16th International Conference on Quality in Research (QIR): International Symposium on Electrical and Computer Engineering, Padang, Indonesia, 2019, pp. 1–5. doi: 10.1109/qir.2019.8898135. [7] M. Pasha, F. Shahzad, and A. Ahmad, “Analysis of challenges faced by WebRTC videoconferencing and a remedial architecture,” International Journal of Computer Science and Information Security, vol. 14, no. 10,pp. 698–705, Oct. 2016. doi: 10.48550/arxiv.1701.09182. [8] K. Sakakibara, S. Ohzahata, and R. Yamamoto, “Deep Learning-Based No-Reference Video Streaming QoE Estimation Using WebRTC Statis-tics,” in 2024 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), Osaka, Japan, 2024, pp. 365–370. doi: 10.1109/icaiic60209.2024.10463278. [9] C.-K. Tsung, H.-Y. Hsieh, and C.-T. Yang, “An Implementation of Scalable High Throughput Data Platform for Logging Semiconductor Testing Results,” IEEE Access, vol. 7, pp. 26497–26506, 2019. doi: 10.1109/access.2019.2901115.A. Panesar, Machine Learning and AI for Healthcare, Apress, 2021. doi: 10.1007/978-1-4842-6537-6
Publication Date
6/14/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
3
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...