In-vivo dosimetry (IVD) is one of the most accurate ways for determining the dosage provided to ‎the patient ‎during treatment‏.‏‎ The aim of this work is to verify whether patients were receiving the ‎accurate prescribed dose. Which occurs by assessing the dose delivery errors,‎‏ ‎caused by human or ‎equipment malfunctioning, using in-vivo diode dosimetry (IVDD).‎‎‏ ‏In total, 302 fields were ‎performed for 80 cases, including pelvis, abdomen, thorax, and head and neck (H & N)‎‏ ‏patients. ‎Prior to the clinical application, Alderson Rando phantom was used to evaluate the diodes' ‎dependability as a complement to the validation process. ‎The results revealed that the measured ‎dose of the ‎phantom ‎was within ±5% of the planned dose‎. Additionally, patients'‎‏ ‏doses achieved ‎‎91.4%‎ of the measured dose were within ±5% of the planned dose. Of the other 8.6%, about 5.6 ‎‎% of the measured doses were more than 5% and less than 10% and ‎only 3% were larger than ‎‎10% of the planned dose.‎ The outlying values that were more than 5% were repeated as in-vitro ‎measurements on a phantom‎ and the deviations were within ‎‎±5%. This study demonstrated that ‎diode measurements provide an immediate readout during the ‎treatment process and it is reliable ‎as quality assurance for linear ‎accelerator. Moreover, ‎IVDD is capable of detecting major and ‎common treatment errors such as patient setup errors and ‎incorrect source surface distance.‎