ABSTRACTCurrently, diagnosis of gas bubble disease (GBD) in fish relies on visual observations of bubbles in various body parts and exophthalmus. These observations lack quantified information about gas bubble load, limiting their diagnostic value. This study evaluated the suitability of existing and novel diagnostic tools for detecting GBD in Atlantic salmon exposed to supersaturation induced by pressure reduction. Existing methods were complemented with semi‐quantitative scoring models of gas bubbles or tissue damages and tested for their potential diagnostic value. Additionally, methods commonly used for diagnosis of GBD in humans, such as detection of gas bubbles in blood using ultrasound, angiotensin converting enzyme (ACE) concentration and haematological parameters, were tested for their suitability for salmon. This study demonstrated that semi‐quantification of gas bubbles using ultrasound was the most reliable method to evaluate the risk of developing acute GBD and to determine whether it is the cause of mortality. The findings suggest that the other assessed methods are unreliable for detecting acute GBD in fish.