Outcome prediction comparison of ischaemic areas’ radiomics in acute anterior circulation non-lacunar infarction
2024

Predicting Outcomes in Acute Anterior Circulation Non-Lacunar Infarction Using MRI Radiomics

Sample size: 372 publication Evidence: high

Author Information

Author(s): Zhou Xiang, Meng Jinxi, Zhang Kangwei, Zheng Hui, Xi Qian, Peng Yifeng, Xu Xiaowen, Gu Jianjun, Xia Qing, Wei Lai, Wang Peijun

Primary Institution: Tongji Hospital, School of Medicine, Tongji University, Shanghai, China

Hypothesis

This study aims to develop and compare machine learning models based on MRI radiomics of multiple ischaemic-related areas for prognostic prediction in acute anterior circulation non-lacunar infarction.

Conclusion

The random forest model with mask ADC620 can accurately predict the AACNLI outcome and reveal the risk factors leading to poor prognosis.

Supporting Evidence

  • The AUC of the model with ADC620 was 0.98 in the internal test set and 0.91 in the external test set.
  • The weighted average AUC in the three sets was 0.96, the largest among three mask types.
  • Severe neurological impairment within 7 days of onset and stroke-associated pneumonia indicates poor prognosis of acute ischaemic stroke.

Takeaway

Doctors can use special pictures of the brain to better guess how well a patient will do after a stroke, helping them give better treatment.

Methodology

This retrospective multicentre study included 372 AACNLI patients and used machine learning models based on MRI radiomics features extracted from various imaging techniques.

Potential Biases

Selection bias may affect the results due to the limited sample size.

Limitations

The sample size is limited, and the study does not include cases treated with thrombolysis or thrombectomy.

Participant Demographics

Patients included were adults aged 18 and older with acute anterior circulation non-lacunar infarction.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1093/braincomms/fcae393

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication