These robots are “allies” of children with autism spectrum disorders

Behavioral therapy robotics is successfully used in several centers: and now more expressive models for children are being studied

The effectiveness of behavioral therapy assisted by robots on children with autism spectrum disorders it is now a full-fledged reality used in numerous assistance centers. IRIB Cnr of Messina has developed several protocols over the years rehabilitation for the treatment of linguistic, visual-spatial and social interaction disorders“As mediated by the robot is able to have greater efficiency in treatment”. But how many and which are the most used robots in the world today to treat children with autism? This question was answered by a recent publication from the group of researchers led by Giovanni Pioggia, head of IRIB Cnr of Messina published in Children magazine.

“The robots on the market today – adds Giovanni Pioggia – however they have some limitations. Our analysis suggests that to design and develop meaningful robot-mediated interventions, the robot must address the needs of children with ASD, healthcare professionals, and developers. The current state of the art of social care therapy has not yet reached its full potential in terms of physical appearance and technological capabilities, which are the two key aspects highlighted in several studies. The robots that have been used so far are being used in a guided way with the help of healthcare professionals. Some robots also have limited mobility capabilities with a visually and kinetically simple design. The challenge for the future is to design a new era of child-friendly expressive humanoid robots to enhance the complex triadic interaction between teachers and children with robots, also considering the introduction of artificial intelligence algorithms, which will thus be able to introduce flexibility and learning capacity by replacing the current rigid conditions. Only a future generation of robots will be able to respond to the different treatment needs of children with autism by increasingly converging towards individualized treatment”.

“In the future, robots – explains researcher Alfio Puglisi research by IRIB Cnr in Messina – will increasingly become part of society and in particular there will be implementations in various areas, from industry to medicine. All dictated by hardware developments, related to sensors that will allow more and more accurate acquisitions from the analog worldto greater computing power that will enable increasingly complex artificial intelligence algorithms to be run on the same algorithms, helping robots to learn behavior from users and adapt to respond in a personalized way according to circumstances.

“At the moment – he also explains the psychologist from IRIB Cnr of Messina Flavia Marino – the robots that are most used and validated in the clinical practice of autism are the models Nao, Kaspar, Qtrobot, Face and Zeno. Behavioral treatments are the most important tool for reducing difficulties in daily life in children with autism spectrum disorder (ASD). Generally these are focused on developing or improving social communication skills and abilities. The past two decades have seen the emergence of technology-based therapies, such as robot-assisted therapy (rat), to improve the treatment of people with Asd. Recent robot-mediated intervention studies have shown positive results in improving joint attention, social communication, imitation and social behavior”.

“Interaction with robots as if they were peers – he adds – is naturally more attractive and motivating because it lowers the level of anxiety, usually generated by social stimuli that people with autism have difficulty decoding. The rat approach has two fundamental advantages: the ability to record objective data during therapy and the ability of the robot to adaptively ‘learn’ both the inter-individual differences at a given time and the intra-individual differences over time, thus partially overcoming limitations due to clinical heterogeneity.The first function is important to characterize behavioral improvement by providing quantitative data on the development process. Thanks to these robots, we have created different protocols for therapy with children that achieve excellent results”.

Artificial intelligence and virtual reality are “forming” tomorrow’s surgeons

Personalized training courses that are able to adapt to the individual student’s learning curve and facilitate the training of future surgeons who will operate with the robot. All this thanks to artificial intelligence and a virtual reality-based simulatoras demonstrated by a study carried out at the University of Pisa, which involved an international research group led by Andrea Moglia from the Department of Translational Research and New Technologies in Medicine and Surgery in collaboration with Professor Sir Alfred Cuschieri from the University of Dundee (United Kingdom).

The study – informs Unipi – required a group of medical students to perform specific exercises on the robotic surgery simulator until the required level of competence was reached. They were later developed via software different artificial intelligence models to predict the weather and the number of trials of each exercise in which the participants would complete their training.

By using a certain configuration (called ensemble learning), where several models of neural networks were aggregated, it was possible to achieve a higher accuracy (up to 95% in some cases) than any other model of machine learning, a branch of artificial intelligence. The results of the study, in which he also participated the general surgery group of Professor Luca Morelli, has been published in Surgical Endoscopy, one of the most authoritative journals in surgery.

The study carried out at the University of Pisa represents the highest international case history in the sector and is based on Andrea Moglia’s experience with simulation in robotic surgery, which began in 2009 at the EndoCas center, led by Professor Mauro Ferrari, and which saw him use, first in Italy, the virtual simulators for the da Vinci robotic surgery system.

“Exactly ten years ago in this period – comments Moglia – I collected data from the survey to evaluate the innate aptitude for surgery among medical students. This time we thought of using artificial intelligence to predict their learning curve with the aim of laying out the training path based on the innate abilities of the learners. Our approach is not limited to the da Vinci robot, but can also be applied to the next generation of robotic surgical systems coming to market. Furthermore, it can be extended to any surgical simulator”.

Artificial intelligence allows computers to learn to perform certain tasks, including trend forecasting, through mathematical models built from data. The availability of increasingly advanced computers and models has meant that in recent years artificial intelligence has been successfully applied to various sectors, primarily self-driving cars. The healthcare system also benefits. One of the last branches of medicine to turn to artificial intelligence is surgery, which potentially offers numerous applications starting with the training of surgeons.

Leave a Comment