Objective To explore the application of rectus femoris cross-sectional area （RF-CSA）， Tibialis anterior muscle thickness （TA-MT）， and bioelectrical-impedance-analysis （BIA） in the diagnosis of intensive care unit-acquired weakness （ICU-AW）. Methods A single-centre， observational study was carried out on 116 patients who were admitted to the Intensive Care Unit （ICU） of Tianjin Medical University Cancer Hospital because of tumour-related sepsis or septic shock. By examining the patients' general clinical data and using ultrasound and bioimpedance analysis to monitor the changes in indicators such as rectus femoris， tibialis anterior， skeletal muscle index （SMI）， total body water （TBW）， and protein content within 6 hours after the onset of sepsis and 72 hours after the initiation of sepsis treatment， statistical analysis was performed to predict the probability of ICU-AW occurrence. Results Ultimately， a total of 41 patients were diagnosed with ICU-AW. Patients with ICU-AW had a higher proportion of vasoactive drug use and mechanical ventilation， along with elevated Acute Physiology and Chronic Health Evaluation Ⅱ （APACHE Ⅱ） and Sequential Organ Failure Assessment （SOFA） scores， which clearly indicated a more severe state of illness. Moreover， group-by-group comparisons showed more significant decreases in RF-CSA， TA-MT， SMI， TBW， and protein content （Protein） in the ICU-AW group （P < 0.05）. In addition， compared to the non-ICU-AW group， ICU-AW patients had lower initial mean Passive Angle （PA） values （3.42° vs. 3.80°） and showed a more rapid increase in PA after aggressive intervention. When it came to the diagnosis of ICU-AW， the change rates of RF-CSA， TA-MT， SMI， TBW， and Protein presented higher Receiver Operating Characteristic-Area Under the Curve （ROC-AUC） values for diagnosing ICU-AW ［0.891 （95%CI： 0.831 ~ 0.952）， 0.830 （95%CI： 0.749 ~ 0.911）， 0.916 （95%CI： 0.862 ~ 0.971）， 0.833 （95%CI： 0.749 ~ 0.917）， and 0.834 （95%CI： 0.758 ~ 0.911）， respectively］. Furthermore， the initial APACHE Ⅱ score showed good diagnostic value， with a ROC-AUC of 0.829 （95%CI： 0.803 ~ 0.917）. In the combined diagnostic model， the joint model that included the SMI change rate and Protein change rate measured by bioimpedance analysis achieved the highest ROC-AUC （0.938， 95%CI 0.895 ~ 0.980）. Conclusion Peripheral muscle ultrasound and bioelectrical impedance analysis can be employed as effective tools for the early detection of ICU-AW in patients with tumor sepsis.