The new leap innovation “Artificial Intelligence” (AI) will significantly accelerate the speed of science and translate into clinical practice faster than clinicians ever have been used to. Machine and deep learning techniques are now readily available to analyze large and complex data sets, providing veterinary scientists with a tool to not only better understand complex behaviour traits and neurological conditions, but also to delve into the nuances of animal behaviour itself. AI can now write plausible sounding text, recognize written or spoken human words, recognize objects in images, or analyze large amounts of data. Different fields of applications are known so far, such as improving and speeding up interpretation of radiologic findings, interpretation of electroencephalography, diagnosis of autism in children, objectifying ataxia in dogs, detecting pain in cats, and assessing and characterizing animal behaviours across species.
At the time of writing, more than 500 AI-based algorithms have been approved by the FDA for human medicine, most of which are used in radiology and neurology. In veterinary neurology and animal behavioural science, only a limited amount of research on AI has been done so far. Most algorithms presented are merely fragments of usable programs and are not suitable for clinical application. Therefore, ready-to-use applications are rare, yet essential for implementing AI into everyday routine diagnostic work-ups, scientific research, and animal behavioural assessments.
The research topic will help to unite current efforts and provide a collection of articles to address some of the current scientific and behavioural gaps. In the future, a clinician might have a clinical decision support tool that generates accurate estimations of likely outcomes of neurological disease and animal behaviour, combining anamnestic, behavioural, clinical, (clinical) pathology, and imaging information.
We are interested in original research, systematic reviews, and opinion papers not only about the scope and potential of AI for Veterinary Behavior and Neurology, but also about the intersection of AI with the study of animal behaviour, ethical considerations, and pitfalls that the development of AI-generated tools could present. Potential themes therefore include, but are not limited to:
• AI to recognize, monitor, diagnose, or characterise behaviour, neurological deficits or conditions, including a detailed analysis of animal behaviour across a variety of contexts and species.
• AI to assess behavioural states in both clinical and natural environments.
• AI to interpret scientific clinical data sets, such as clinical pathology, omics, imaging, EEG data, and behaviour metrics.
• Ethical and legal considerations when applying or using AI tools in the study of animal behaviour and veterinary medicine.
• Potential of AI to enhance clinical practice in veterinary behavioural medicine, veterinary neurology, and animal behaviour research.
• Potential of AI-enhanced diagnostic and behavioural assessment tools with animals.
The new leap innovation “Artificial Intelligence” (AI) will significantly accelerate the speed of science and translate into clinical practice faster than clinicians ever have been used to. Machine and deep learning techniques are now readily available to analyze large and complex data sets, providing veterinary scientists with a tool to not only better understand complex behaviour traits and neurological conditions, but also to delve into the nuances of animal behaviour itself. AI can now write plausible sounding text, recognize written or spoken human words, recognize objects in images, or analyze large amounts of data. Different fields of applications are known so far, such as improving and speeding up interpretation of radiologic findings, interpretation of electroencephalography, diagnosis of autism in children, objectifying ataxia in dogs, detecting pain in cats, and assessing and characterizing animal behaviours across species.
At the time of writing, more than 500 AI-based algorithms have been approved by the FDA for human medicine, most of which are used in radiology and neurology. In veterinary neurology and animal behavioural science, only a limited amount of research on AI has been done so far. Most algorithms presented are merely fragments of usable programs and are not suitable for clinical application. Therefore, ready-to-use applications are rare, yet essential for implementing AI into everyday routine diagnostic work-ups, scientific research, and animal behavioural assessments.
The research topic will help to unite current efforts and provide a collection of articles to address some of the current scientific and behavioural gaps. In the future, a clinician might have a clinical decision support tool that generates accurate estimations of likely outcomes of neurological disease and animal behaviour, combining anamnestic, behavioural, clinical, (clinical) pathology, and imaging information.
We are interested in original research, systematic reviews, and opinion papers not only about the scope and potential of AI for Veterinary Behavior and Neurology, but also about the intersection of AI with the study of animal behaviour, ethical considerations, and pitfalls that the development of AI-generated tools could present. Potential themes therefore include, but are not limited to:
• AI to recognize, monitor, diagnose, or characterise behaviour, neurological deficits or conditions, including a detailed analysis of animal behaviour across a variety of contexts and species.
• AI to assess behavioural states in both clinical and natural environments.
• AI to interpret scientific clinical data sets, such as clinical pathology, omics, imaging, EEG data, and behaviour metrics.
• Ethical and legal considerations when applying or using AI tools in the study of animal behaviour and veterinary medicine.
• Potential of AI to enhance clinical practice in veterinary behavioural medicine, veterinary neurology, and animal behaviour research.
• Potential of AI-enhanced diagnostic and behavioural assessment tools with animals.