QuanTiMedAI: Quantum-Enhanced Time-Series Model guided by Agentic AI for Cardiac Arrest Mortality Prediction

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Mutasim Fuad Sarker, Adiba Rahman Namira, Wafa Binte Alam, Md Adnan Arefeen, Mahzabeen Emu, Sumaiya Tabassum Nimi
• Venue: arXiv: cs.AI
• Published: 2026-08-06
• Collected: 2026-08-07
• URL: http://arxiv.org/abs/2608.06294v1
• PDF: https://arxiv.org/pdf/2608.06294v1
• DOI: UNKNOWN

Plain-Language Brief

Researchers introduce QuanTiMedAI, a framework for predicting cardiac arrest mortality using a combination of an agentic large language model (LLM) and a quantum recurrent network. The LLM is used for feature discovery, while the quantum network is used for temporal awareness in mortality prediction. The proposed system outperforms conventional approaches, achieving an AUROC of 0.852 with significantly fewer parameters.

What the Document Says

• Cardiac arrest remains a severe condition in ICUs and existing mortality prediction studies often rely on early admission summaries.
• A new framework, QuanTiMedAI, is introduced that combines an agentic LLM for feature discovery with a quantum recurrent network for temporal awareness.
• The system uses the LLM to select features consistently better than conventional methods and the quantum architecture achieves competitive performance with nonlinear feature enhancement using few parameters.
• In experiments on a MIMIC-IV cohort, QuanTiMedAI's quantum-enhanced architecture outperforms a baseline by approximately 2.9% with significantly fewer parameters.
• A structured ablation study verifies the impact of each architectural design choice.

Key Concepts Mentioned

• Cardiac arrest
• ICU
• electronic health record data
• mortality prediction
• Quantum-Enhanced Time-Series Model
• Agentic AI
• Large Language Model (LLM)
• quantum recurrent network
• AUROC
• MIMIC-IV cohort
• ablation study

Claims Made by the Document

• Existing mortality prediction studies largely rely on early admission summaries and ignore the temporal progression of physiological deterioration.
• QuanTiMedAI outperforms conventional feature selection approaches.
• The proposed quantum architecture achieves competitive predictive performance with fewer parameters.
• QuanTiMedAI's quantum-enhanced architecture outperforms a baseline by approximately 2.9%.

Evidence Used

• The text states that existing studies rely on early admission summaries and ignore the temporal progression of physiological deterioration.
• The text mentions that QuanTiMedAI outperforms conventional feature selection approaches.
• The text states that the proposed quantum architecture achieves competitive performance with fewer parameters.
• The text mentions that QuanTiMedAI's quantum-enhanced architecture outperforms a baseline by approximately 2.9%.

Methods / Approach

• QuanTiMedAI combines an agentic LLM for feature discovery and a quantum recurrent network for temporal awareness.
• The system uses an ablation study to validate the impact of each architectural design choice.

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

The source does not provide information on potential limitations of the proposed method beyond those mentioned by the authors.

Open Questions

• What is the specific nature of the ablation study conducted?
• What is the nature of the quantum recurrent network used in QuanTiMedAI?

Notes for Future Q&A

May include interpretation. Labeled where it does.

• It would be beneficial to understand the details of the ablation study and the quantum recurrent network.
• Further research may investigate potential limitations of the proposed method not mentioned by the authors.

Raw Abstract / Excerpt

Cardiac arrest remains one of the most lethal conditions encountered in intensive care units. Despite the growing availability of electronic health record data, existing mortality prediction studies in this population largely depend on static summaries derived from early admission. Such approaches ignore the temporal progression of physiological deterioration and recovery that unfolds throughout a patient's ICU stay. To address this limitation, we introduce QuanTiMedAI, a quantum-agentic framework developed for cardiac arrest mortality prediction using agentic AI guided quantum enhancement time series model. The proposed system combines an agentic large language model (LLM) for clinically informed feature discovery with a compact quantum recurrent network for temporality aware mortality prediction. Our findings demonstrate that agentic LLM-guided feature selection consistently outperforms conventional feature selection approaches, and the proposed quantum architecture achieves competitive predictive performance through nonlinear feature enhancement while keeping the number of parameters very low. Through extensive experimentation on a MIMIC-IV cohort of cardiac arrest patients, QuanTiMedAI's quantum-enhanced architecture attains an AUROC of 0.852 using only 605 parameters, an improvement of approximately 2.9% over a current state-of-the-art baseline for this task. A structured ablation study systematically validates the contribution of each architectural design choice. These results show that quantum-enhanced sequential modeling can exceed classical recurrent networks while using substantially fewer parameters.