The research team's aim is to investigate the effects of the team's soft robotic socks in providing assisted ankle dorsiflexion-plantarflexion and promoting venous blood flow in stroke patients. Specifically, the investigators intend to 1) observe the development of joint contracture in terms of the range of motion of the ankle joint given the use of the device 2) monitor the presence of blood clots in the deep veins using compression Duplex ultrasonography given the use of the sock device.

The investigators hypothesize that a soft robotics approach can provide compliant actuation to simulate natural ankle dorsiflexion and plantarflexion, which will consequently promote blood circulation in the lower leg of stroke patients who have not previously developed ankle joint contracture. This will prevent the occurrence of DVT and movement in the ankle joint will alleviate stiffness. This robotic sock intervention is to prevent the development of ankle joint contracture so established ankle contractures will be excluded from this study.

The investigator's proposed robotic sock solution is capable of providing programmable robot-assisted ankle exercises to chronic bedridden patients, thereby improving venous blood flow, and preventing DVT and ankle joint contractures. Compared to conventional mechanical prophylaxis such as the intermittent pneumatic compression devices, the robotic sock is likely to cost almost five times less, and potentially generate patient/hospital savings of up to $2750 per patient. The use of the robotic sock will also raise therapist productivity through automating the therapy exercises of the bedridden patients, and this could potentially save the therapists at least 6 workhours per day. The use of the robotic sock in hospitals, nursing homes and patients' own homes will likely reduce healthcare costs and prevent side effects, as compared to conventional DVT prophylaxis approaches. In addition, it will add new values and benefits to public healthcare by eliminating additional treatment costs arising from DVT-related complications, increasing therapists' productivity (especially given growing manpower constraints and greying population), optimizing therapy time, and ultimately saving precious lives.

Among those diagnosed with DVT, 10-30% dies within 1 month of diagnosis (Beckman et al., 2010), and worldwide 600,000-800,000 people die annually (worldthrombosisday.org). This research seeks to lower DVT risk through a safe non-drug-based approach by continuously moving the ankle joint into dorsiflexion-plantarflexion using soft robotics. Current alternatives to prevent DVT include pharmacological prophylaxis such as anticoagulant drugs, mechanical prophylaxis such as intermittent pneumatic compression devices and compression stockings, and the basic therapist-assisted exercises to prevent ankle joint contractures. Pharmacological prophylaxis can improve venous blood flow and prevent DVT, but run the risk of detrimental side effects like excessive bleeding. Mechanical prophylaxis can provide passive or automated stimulation to the calf tissue, but have been reported to deliver limited efficacy in improving venous blood flow and preventing DVT. Considering the data from the United States and United Kingdom, the annual spending incurred due to direct and indirect costs resulting from DVT is approximated to be US$2-10 billion and this ranges from US$7,594-US$16,644 per patient (Spyropoulos and Lin, 2007). Therefore, by implementing the robotic sock device, the investigators hope to enable clinicians to focus on treating patients towards stroke recovery without worrying about other complications.

Joint contractures, usually defined as limited passive range of joint motion are common in people with neurological conditions such as stroke or spinal cord injury. The risk factors for joint contractures are not well understood; however, immobility seems to be the most important factors. Joint contractures, especially in the ankle joints, may impede the ability to walk and cause loss of balance, high risk of falls and restricted participation in social activities. The frequency of ankle contractures in hospitalized patients varies with prevalence rates ranging from 24 % to 44 %.

Inclusion Criteria:

- Ischemic or haemorrhagic stroke patients with severe lower limb weakness (MRC scale <3) in flexors and extensors of knee and ankle

Exclusion Criteria:

• Medically unstable patients
• Queried pulmonary embolism (PE) / deep vein thrombosis (DVT) cases
• Limited range of motion in ankle and foot including equinus or club foot deformity
• Lower limb dermatitis, ulcer or open wound
• Severe spasticity of ankle (modified Ashworth scale >2)
• Pregnancy