Development of a Smart Glove for Remote Monitoring of Paediatric Limb Deformities

全世界每年约有20万名新生儿患有手部和上肢畸形。这些都是由广泛的医疗条件（如脑瘫），并需要医疗和手术干预。对精细运动技能的可靠评估对于评估这些干预措施和监测症状至关重要。客观地评估儿童的手和上肢功能是一个挑战。沟通困难意味着无法使用标准方法。记录患者的准确关节位置，无需培训，可以准确地显示信息，不会干扰结果，这将为诊断和评估患者提供宝贵的工具，特别是在可能无法面对面互动的情况下;就像2020年COVID-19大流行一样。该项目旨在开发一种智能手套，当患者通过与临床医生的电话会议完成任务时，读取和记录手部运动和抓握特征。将开发软件来帮助临床医生解释数据。该软件将被设计为将测量结果映射到临床相关参数，并提供对患者手部运动的更详细了解。使用该软件，临床医生将能够远程进行诊断，并客观地评估治疗的成功。手套的设计将确保在收集数据时不会干扰患者，并且足够简单，可以在自然的家庭环境中使用。最终的设计将在伯明翰儿童医院的上肢和手部畸形儿童身上进行评估。快速识别模式使机器学习在简化诊断过程中发挥关键作用，自动检测关键特征，并突出显示准备好供临床医生检查的区域。它还将促进更快和更有信心的诊断和患者状态的分类。患有手部和上肢畸形的儿童通常无法充分沟通，从而使诊断变得困难。不正确的诊断和低质量的评估可能导致无效的手术，导致进一步的纠正程序。我们的手套的好处将包括显着改善患者的功能信息;节省诊所和临床医生的时间;访问专家单位的次数;限制患者的痛苦;并最终提高生活质量和功能。

Every year approximately 200,000 children are born worldwide with hand and upper limb deformities. These are caused by a wide range of medical conditions (e.g. cerebral palsy) and require medical and surgical interventions. A reliable assessment of fine motor skills is crucial for evaluating these interventions and monitoring the symptoms over time. Objectively assessing hand and upper limb function in children is a challenge. Communication difficulties mean standard methods cannot be used. Recording accurate joint positions from the patient in a way that requires no training, can surface information accurately and will not interfere with results will provide an invaluable tool to diagnose and evaluate patients, especially in situations where face to face interactions may not be possible; as with the 2020 COVID-19 pandemic. This project aims to develop a smart glove, to read and record hand motion and grip characteristics as the patient completes tasks over teleconference with a clinician. Software will be developed to help clinicians interpret the data. The software will be designed to map measurements to clinically relevant parameters and provide more detailed insight into patient's hand movements. Using the software clinicians will be able to remotely perform diagnosis and assess the success of treatments objectively. The glove will be designed as not to interfere with the patient while data is being collected and be simple enough to be used in a natural home environment. The final design will then be assessed on children with upper limb and hand deformities at Birmingham Children's Hospital. Opportunities for commercialisation of the device will also be explored.Recognising patterns quickly gives machine learning a key role in streamlining the diagnosis process by detecting key features automatically and highlighting those areas ready for the clinician to review. It will also facilitate faster and more confident diagnosis and classification of a patient's state. Children with hand and upper limb deformities are generally unable to adequately communicate making diagnosis difficult. Incorrect diagnosis and poor-quality assessments can result in ineffective surgery, leading to further corrective procedures. The benefits of our glove will include significant improvement in information on patient's functioning; savings in clinic and clinician's time; number of visits to specialists' units; limiting distress to patients; and ultimately improved quality of life and functioning.