Aim: Diagnostic arthroscopy is an invasive and an expensive method using for the diagnosis of meniscal tears. The aim of this study was to determine the value of knee magnetic resonance imaging in the diagnosis of meniscal tears and its role in the prevention of unnecessary diagnostic arthroscopy. Methods: A total of 105 patients who underwent knee magnetic resonance imaging and arthroscopy due to meniscus injury were included in the study. Fifty-nine patients were examined using a 1.5 Tesla magnetic resonance scanner and 46 were examined using a 3.0 Tesla magnetic resonance scanner. Magnetic resonance imaging findings were evaluated retrospectively in workstations by two radiologists experienced in musculoskeletal magnetic resonance imaging. Meniscal tears were reported as anterior horn tear, corpus tear or posterior horn tear. Meniscal tears were classified by using surgical classification. Each patient's magnetic resonance images were evaluated with a consensus and compared with the arthroscopic diagnosis. Results: Meniscal tears were detected in 96 out of 106 knees on arthroscopy. By using arthroscopy as the gold standard for diagnosis of meniscal tears, the sensitivity, specificity, and accuracy values of the magnetic resonance imaging evaluation were found as 85.71% (95% CI: 77.84-91.61), 93% (95% CI: 86.11-97.14) and 89.15% (95% CI: 84.17-93), respectively. These values of magnetic resonance imaging were found high in the diagnosis of meniscal tears. Conclusions: Magnetic resonance imaging is an effective imaging method in the diagnosis of meniscal tears. Although the resolution of 3.0 Tesla magnetic resonance imaging is higher, 1.5 Tesla magnetic resonance imaging is sufficient in routine meniscus tear diagnosis.