mtslearn: Machine Learning in Python for Medical Time Series

Zhongheng Jiang; Yuechao Zhao; Donglin Xie; Chenxi Sun; Rongchen Lu; Silu Luo; Zisheng Liang; Shenda Hong

cs.LGMar 31, 2026

mtslearn: Machine Learning in Python for Medical Time Series

Authors: Zhongheng Jiang, Yuechao Zhao, Donglin Xie, Chenxi Sun, Rongchen Lu, Silu Luo, Zisheng Liang, Shenda Hong

Organizations: Nanjing University of Information Science and Technology, Nanjing, China · National Institute of Health Data Science, Peking University, Beijing, China · School of Health Humanities, Peking University Health Science Center, Beijing, China · Institute of Medical Technology, Peking University Health Science Center, Beijing, China · Harvard Medical School, Boston, US · Beth Israel Deaconess Medical Center, Boston, US · School of Business English, Sichuan International Studies University, Chongqing, China · School of Clinical Medicine, Chengdu Medical College, Chengdu, Sichuan, China · Institute for Artificial Intelligence, Peking University, Beijing, China · State Key Laboratory of Vascular Homeostasis and Remodeling, NHC Key Laboratory of Cardiovascular Molecular Biology and Regulatory Peptides, Peking University, Beijing, China

Abstract

Medical time-series data captures the dynamic progression of patient conditions, playing a vital role in modern clinical decision support systems. However, real-world clinical data is highly heterogeneous and inconsistently formatted. Furthermore, existing machine learning tools often have steep learning curves and fragmented workflows. Consequently, a significant gap remains between cutting-edge AI technologies and clinical application. To address this, we introduce mtslearn, an end-to-end integrated toolkit specifically designed for medical time-series data. First, the framework provides a unified data interface that automates the parsing and alignment of wide, long, and flat data formats. This design significantly reduces data cleaning overhead. Building on this, mtslearn provides a complete pipeline from data reading and feature engineering to model training and result visualization. Furthermore, it offers flexible interfaces for custom algorithms. Through a modular design, mtslearn simplifies complex data engineering tasks into a few lines of code. This significantly lowers the barrier to entry for clinicians with limited programming experience, empowering them to focus more on exploring medical hypotheses and accelerating the translation of advanced algorithms into real-world clinical practice. mtslearn is publicly available at https://github.com/PKUDigitalHealth/mtslearn.

Figures & tables

Figure 1: Evolutionary paradigm of clinical prediction modeling. Rather than introducing complex deep learning algorithms for time-series data, this study builds upon traditional statistical methods and classic deep learning algorithms. We integrated established AI medical tools into a unified framework to simplify workflows for clinicians.

Figure 2: Structural representations of heterogeneous clinical data. The schematic illustrates the three most common tabular formats encountered in clinical records: Wide format, Long format, and Flattened format. mtslearn standardizes these diverse inputs into a unified representation to significantly reduce manual preprocessing overhead.

Name	Preprocessing			Predict		End-to-End
Name	Data cleaning	Time-series resampling	Feature Stats	Static model	Time-series model	Pipelining
TemporAI 19	$\checkmark$	$\times$	$\times$	$\checkmark$	$\checkmark$	$\checkmark$
Clairvoyance 8	$\checkmark$	$\checkmark$	$\times$	$\checkmark$	$\checkmark$	$\checkmark$
NeuroKit2 14	$\checkmark$	$\checkmark$	$\checkmark$	$\times$	$\times$	$\times$
Talent 12	$\checkmark$	$\times$	$\times$	$\checkmark$	$\times$	$\checkmark$
scikit-learn 17	$\checkmark$	$\times$	$\times$	$\checkmark$	$\times$	$\checkmark$

Table 1: Functional comparison of existing machine learning toolkits versus mtslearn. The table highlights capabilities across data preprocessing and model prediction, which shows that most existing toolkits focus on specific stages. ("End-to-End Pipeline" denotes a unified workflow integrating the entire lifecycle, from raw data processing to model evaluation.)

Figure 3: Overview of the mtslearn pipeline.

Figure 4: Before and after data reading.

MedMamba: Multi-View State Space Models with Adaptive Graph Learning for Medical Time Series Classification
May 24, 2026Da Zhang, Bingyu Li, Zhiyuan Zhao +3Graph Structure Learning Selective SSMs
Northwest Polytechnical University · TeleAI, China Telecom · TeleAI +1
MedTS-TTT: Test-Time Training for Medical Time Series Classification
Jun 19, 2026Mingzhi Chen, Yiyu Gui, Guibo LuoTime Series Classification Electrocardiogram Classification
Guangdong Provincial Key Laboratory of Ultra High Definition Immersive Media Technology, Shenzhen Graduate School, Peking University, Shenzhen, China
DoctorAgents: an agentic framework to iteratively refine AutoML pipeline for small clinical temporal data
Aug 5, 2026Ruilin Wang, Bo-Hong Wang, Elizabeth Kourbatski +6LLM Agent Orchestration Automated Machine Learning
School of Computer Science, McGill University, Montreal, Canada · Mila – Quebec AI Institute, Montreal, Canada · Centre intégré universitaire de santé et de services sociaux de l’Estrie – Centre hospitalier universitaire de Sherbrooke (CIUSSSE-CHUS), Sherbrooke, Canada +2

May 24, 2026J/K move · Enter open · S save

MedMamba: Multi-View State Space Models with Adaptive Graph Learning for Medical Time Series Classification

Da Zhang, Bingyu Li, Zhiyuan Zhao, Hongyuan Zhang+2

Northwest Polytechnical University · TeleAI, China Telecom · TeleAI+1

Graph Structure Learning Selective SSMs Multivariate Time Series Classification Healthcare Time Series Classification Non-Stationary Time Series+1

Open paper AskPDF arXiv

Medical time series are central to healthcare, enabling continuous monitoring and supporting timely clinical decisions. Despite recent progress, existing methods struggle to jointly model local-global dynamics and handle nonstationarities like baseline drift, while often failing to capture latent channel interactions. To address these challenges, we propose MedMamba, an end-to-end architecture that integrates state space models with domain-specific inductive biases. Specifically, MedMamba first employs multi-scale convolutional embeddings to capture discriminative local morphology. Second, to mitigate nonstationarity, we introduce a tri-branch differential state space encoder that processes raw, temporal-difference, and frequency-domain views, fusing them to emphasize informative patterns while suppressing drift. Furthermore, to uncover latent channel correlations, we design a spatial graph Mamba module that learns a directed dependency structure regularized toward sparsity and acyclicity, which obviates the need for predefined graphs. Extensive experiments on five real-world datasets demonstrate that MedMamba achieves state-of-the-art performance while maintaining linear computational complexity, and ablation studies validate each component's contribution.Code is available at https://github.com/zhangda1018/MedMamba.

Abstract

Explore similar work

MedMamba: Multi-View State Space Models with Adaptive Graph Learning for Medical Time Series Classification

MedTS-TTT: Test-Time Training for Medical Time Series Classification

DoctorAgents: an agentic framework to iteratively refine AutoML pipeline for small clinical temporal data