With the application of artificial intelligence (AI) and end-to-end recognition methods in handwritten mathematical expression recognition, there has been a significant improvement in recognition accuracy. However, in contrast to tests on public datasets, real-world applications involving human input introduce more uncertain factors into recognition algorithms in practice. Factors such as personalized stroke information, ambiguous handwritten characters, and uncertain formula structures can significantly impact the performance of the recognition method. To address these challenges, HchMER, a hybrid human-machine intelligence method for handwritten mathematical expression recognition, is proposed. HchMER combines handwritten mathematical formula recognition algorithms, knowledge bases, and user feedback to enhance the machine's comprehension of user-input mathematical expressions, thereby improving the editing speed and accuracy of handwritten mathematical expressions. To assess the effectiveness of HchMER, it is compared with MyScript Math Recognition (MyScript) and a mature commercial product named “Microsoft Ink Equation” (InkEquation). Results show that HchMER outperformed MyScript and InkEquation in accuracy by 23.2% and 26.51%, respectively. In terms of average completion time, HchMER exceeded MyScript by 44.46% (9.6s) but fell short of InkEquation by 11.48% (4.05s). Furthermore, participants affirm HchMER in a questionnaire survey and semi-structured interviews.