A machine learning method correlating pulse pressure wave data with pregnancy

Q: What is the clinical evidence from this study?

Study design: Cross-Sectional. Population: Pregnancy (n=1840). Intervention: Deep learning analysis of pulse pressure waves vs. Standard pregnancy tests (clinical ground truth). Primary outcome: Accuracy of pregnancy detection (AUC 91%).

Jianhong Chen; Huang Huang; Wenrui Hao; Jinchao Xu

doi:10.48550/arxiv.1910.01726

October 3, 2019Open Access

A machine learning method correlating pulse pressure wave data with pregnancy

View Full Paper

Ask AI

Bookmark

Share

Key result

Deep learning analysis of pulse pressure wave data detected pregnancy with an accuracy of 84% and an AUC of 91%.

Why the study?

The quantitative relationship between pulse wave and health conditions has not been investigated in modern medicine.

Does deep learning analysis of pulse pressure wave data accurately detect pregnancy in women?

Study Design

Type

Cross-Sectional (n=1,840)

Multicenter

No

Structured PICO

Does deep learning analysis of pulse pressure wave data accurately detect pregnancy in women?

P

Population

1,840 women (645 pregnant and 1,195 non-pregnant) with regular pulse pressure wave data collected at a single center in China.

I

Intervention

Deep learning models (1D Convolutional Neural Networks: ResNet-based Model 1 and MgNet-based Model 2) analyzing pulse pressure wave (PPW) data.

C

Comparator

Support Vector Machine (SVM), logistic regression, and other CNN models.

O

Outcome

Accuracy and Area Under the Curve (AUC) for pregnancy detection.

Main Result

Effect estimate: AUC 91%

Deep learning models applied to pulse pressure wave data can detect pregnancy with approximately 85% accuracy, providing a proof of concept for modernizing traditional pulse diagnosis.

Limitations

Measurement noise introduced during collection due to crude pulse acquisition equipment
Dataset size may not be large enough to train a highly accurate deep learning model
The machine learning approach cannot explain the underlying biological mechanisms
data noise due to straightforward and crude pulse acquisition equipment
dataset size of about 4,000 samples is not quite enough to train a high accuracy deep learning model

Implication

May support pulse diagnostics in pregnancy; hypothesis-generating and requires prospective validation before any clinical use.

Abstract

Pulse feeling, representing the tactile arterial palpation of the heartbeat, has been widely used in traditional Chinese medicine (TCM) to diagnose various diseases. The quantitative relationship between the pulse wave and health conditions however has not been investigated in modern medicine. In this paper, we explored the correlation between pulse pressure wave (PPW), rather than the pulse key features in TCM, and pregnancy by using deep learning technology. This computational approach shows that the accuracy of pregnancy detection by the PPW is 84% with an AUC of 91%. Our study is a proof of concept of pulse diagnosis and will also motivate further sophisticated investigations on pulse waves.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Chen et al. (2019) conducted a cross-sectional in Pregnancy (n=1,840). Deep learning analysis of pulse pressure waves vs. Standard pregnancy tests (clinical ground truth) was evaluated on Accuracy of pregnancy detection (AUC 91%). Deep learning analysis of pulse pressure wave data detected pregnancy with an accuracy of 84% and an AUC of 91%.

synapsesocial.com/papers/6a9a3cfcdb3548d93776b39f https://doi.org/10.48550/arxiv.1910.01726

View Full Paper

Ask AI

Bookmark

Share

A machine learning method correlating pulse pressure wave data with pregnancy

Study Design

Structured PICO

Main Result

Limitations

Implication

Abstract

Discussion

Cite This Study

A machine learning method correlating pulse pressure wave data with pregnancy

Study Design

Structured PICO

Main Result

Limitations

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading