<p>Patients undergoing maintenance hemodialysis (MHD) often suffer from sarcopenia, which affects their balance and significantly increases the risk of falls and death. Actively identifying sarcopenia, understanding the relationship between sarcopenia and the distribution of plantar pressure and balance ability in MHD patients, and reducing the risk of unstable falls are crucial. This study aims to develop an auxiliary screening model for sarcopenia in MHD patients based on machine learning methods using plantar pressure biomechanical indicators, while also paying attention to the gender difference and exploring its value in screening for sarcopenia. This study was a cross-sectional study. Data were collected from MHD patients receiving care at a hemodialysis center in China. The Asian Working Group for Sarcopenia consensus updated in 2019 was used to assess whether MHD patients had sarcopenia. Plantar pressure distribution was measured using a plantar pressure plate for MHD patients with their feet-together and separated by 10&#xa0;cm. Patients were grouped into the sarcopenia and control group and grouped into three gender groups (mixed-gender group, male group, and female group). Comprehensive physiological indicators, including the patient’s basic information, body measurements, and laboratory test results, were also collected to enhance model accuracy. After statistical analysis and feature selection, stratified shuffling split and synthetic minority oversampling technique were used to develop auxiliary screening classifiers for sarcopenia. A total of 184 MHD patients’ data were collected, including 95 males (40 with possible sarcopenia or sarcopenia) and 89 females (57 with possible sarcopenia or sarcopenia). Three binary classification models were developed for the mixed-gender group (mean age of 53.66), male group (mean age of 53.43), and female group (mean age of 53.91), respectively, using plantar pressure biomechanical indicators. The area under the receiver operating characteristic curve (AUC) values for the three models were 75%, 72%, and 82%, respectively. The use of plantar pressure indicators in the feet-together stance, when combined with comprehensive physiological indicators (age and pre-dialysis creatinine), further improved the accuracy of the auxiliary screening model (AUC values of up to 78% and 88%, respectively). Plantar pressure indicators have high value in the screening of sarcopenia in MHD patients. When using plantar pressure indicators of the feet-together and feet-separated by 10&#xa0;cm stance, a mixed-gender model should be chosen to obtain better screening results. In addition, if clinical conditions allow it, more accurate results for the screening of sarcopenia can be obtained by combining comprehensive physiological indicators with plantar pressure indicators of the feet-together stance.</p><p><i>Trial registration</i>: Chinese Clinical Trial Registry (ChiCTR2100051111), registered on 20,210,913.</p>