GAIT TRAINING, PRE-AMBULATION PROGRAMME, ASSISTIVE DEVICES AND GAIT PATTERNS, RECENT ADVANCES IN GAIT ANALYSIS

1. GAIT TRAINING

Definition

Goals of Gait Training

• Restore a safe, efficient, and functional ambulatory pattern
• Reduce fall risk
• Improve lower limb strength, coordination, and proprioception
• Transition the patient from assisted to independent ambulation
• Educate the patient on correct use of assistive devices

Candidates

2. PRE-AMBULATION PROGRAMME

Goals

• Improve muscular strength (especially hip extensors, knee extensors, and ankle plantar/dorsiflexors)
• Develop balance and postural control (static and dynamic)
• Build cardiovascular tolerance for upright activity
• Teach correct weight-bearing postures
• Familiarise the patient with assistive devices

Phases of Pre-Ambulation Programme

Phase 1 - Bed Exercises

• Strengthening: Ankle pumps, quadriceps sets (isometric), straight-leg raises, hip abductor and adductor strengthening, bridging exercises
• Range of motion: Passive, active-assisted, and active ROM for hip, knee, and ankle
• Breathing exercises: Diaphragmatic breathing to maintain cardiorespiratory fitness

Phase 2 - Sitting Balance and Transfers

• Static sitting balance: sitting on edge of bed without support
• Dynamic sitting balance: reaching activities, perturbation training
• Sit-to-stand and stand-to-sit transfers (key prerequisite for ambulation)
• Controlled transfer from bed to chair, chair to commode

Phase 3 - Standing Balance at Parallel Bars

• Static standing: patient holds parallel bars, progresses to one-hand hold, then no hands
• Weight shifting: side-to-side and forward-backward shifts
• Step-ups and step-downs in place (marching)
• Single-leg stance: builds unilateral weight-bearing ability
• Parallel bars provide maximal security; the patient learns to walk in a controlled, supported environment before progressing to assistive devices

Phase 4 - Parallel Bar Ambulation

• Forward stepping inside the parallel bars
• Therapist guards with gait belt at all times
• Instruction on step length, cadence, heel strike, and reciprocal arm swing

Phase 5 - Progression to Assistive Devices

Phase 6 - Advanced Gait Training

• Ambulation on uneven surfaces, ramps, and stairs
• Negotiating curbs, doorways, and elevators
• Outdoor ambulation
• Fall-recovery techniques

3. ASSISTIVE DEVICES FOR AMBULATION

Weight-Bearing Categories (Critical for Device Selection)

Classification of Assistive Devices

A. Canes

• Standard (straight) cane: Single point of contact; FWB or WBAT status
• Tripod cane (three-legged): Broader base; for patients needing more lateral stability
• Quad cane (four-legged): Maximum base support among canes; suits hemiplegia or severe balance disorders

B. Axillary (Underarm) Crutches

• Axillary pad: 2-3 finger-widths (5 cm) below axilla
• Handgrip: level of the greater trochanter / wrist crease
• Elbow: 20-30° flexion
• Tips: 15 cm lateral to and 15 cm anterior to the foot

C. Lofstrand (Forearm / Elbow) Crutches

• Forearm cuff wraps around the forearm; handgrip below
• Allow hands to be freed without dropping the crutch (useful for patients needing to use hands briefly)
• Used for long-term ambulation (e.g. paraplegia from polio, incomplete SCI)
• Less stability than axillary crutches

D. Walkers

• Standard walker (pick-up walker): Lifted and placed with each step; maximum stability; used for NWB or PWB; requires good upper limb strength
• Rolling walker (wheeled, 2-wheeled): Front wheels roll; less energy expenditure; used for WBAT or FWB patients with limited endurance (e.g. Parkinson's, elderly)
• 4-wheeled walker (rollator): All four wheels roll; usually has brakes and a seat; used for patients needing to pause and rest; less stable than standard walker
• Hemi-walker: One-sided walker for hemiplegic patients; provides wider base than a quad cane

E. Parallel Bars

4. GAIT PATTERNS (Gait Sequencing Patterns)

1. Four-Point Gait

• Devices: Bilateral crutches or bilateral canes
• Weight-bearing: FWB or PWB bilaterally
• Sequence: Right crutch → Left foot → Left crutch → Right foot
• Characteristics: Most stable pattern (three points of contact always on the floor); very slow; mimics a natural reciprocal pattern
• Indication: Weakness in both lower limbs but sufficient weight-bearing; post-polio, incomplete SCI, MS

2. Two-Point Gait

• Devices: Bilateral crutches or bilateral canes
• Weight-bearing: FWB, PWB, TTWB, WBAT
• Sequence: Right crutch + Left leg simultaneously → Left crutch + Right leg simultaneously
• Characteristics: Faster than four-point; maintains reciprocal arm-leg pattern; requires better balance and coordination
• Indication: Bilateral lower limb weakness with adequate balance

3. Three-Point Gait

• Devices: Bilateral crutches or walker
• Weight-bearing: NWB, TTWB, PWB, WBAT on one limb
• Sequence: Both crutches + involved limb together → uninvolved limb advances
• Characteristics: The involved limb and both crutches advance as a unit; the uninvolved limb then hops or steps forward; requires strong upper limbs and good single-leg balance
• Indication: Unilateral NWB (e.g. fracture, post-operatively); unilateral amputation pre-prosthetically

4. Modified Three-Point Gait

• Devices: Walker or bilateral crutches
• Weight-bearing: WBAT or FWB (where full weight is allowed but patient needs support)
• Sequence: Walker/crutches → involved limb → uninvolved limb (or walker → both limbs together)
• Characteristics: Slower and more conservative; the involved limb advances with the device but full or as-tolerated weight is accepted; energy-saving pattern for patients with reduced endurance
• Note: A PWB patient cannot use modified three-point because PWB requires the involved limb to advance with the device, which makes it a three-point pattern

5. Swing-To Gait

• Devices: Bilateral crutches or walker
• Weight-bearing: NWB bilaterally, or severe bilateral lower limb involvement
• Sequence: Both crutches advance → both feet swing forward to the level of the crutches
• Characteristics: Feet land at or behind the crutches; more stable than swing-through; used when the patient cannot achieve reciprocal stepping
• Indication: Thoracic or high lumbar SCI with bilateral flaccid lower limbs

6. Swing-Through Gait

• Devices: Bilateral crutches
• Weight-bearing: Bilateral NWB (complete lower limb paralysis)
• Sequence: Both crutches advance → both feet swing past the crutches
• Characteristics: Fastest non-reciprocal pattern; requires excellent upper limb strength and trunk balance; high energy demand; greatest risk of falling
• Indication: Paraplegic patients using hip-knee-ankle-foot orthoses (HKAFOs); athletes with spinal cord injuries

Stair Climbing Techniques

• Going up: "Good goes up first" - uninvolved limb leads, then involved limb and crutches
• Going down: "Bad goes down first" - crutches and involved limb descend first, then uninvolved
• Mnemonic: "Up with the good, down with the bad"

5. RECENT ADVANCES IN GAIT ANALYSIS

A. Instrumented Gait Analysis (3D Motion Capture) - Standard of Practice

B. Wearable Inertial Measurement Units (IMUs)

• Continuous, real-world ambulation monitoring (not constrained to lab)
• Assessment of spatiotemporal parameters: stride length, step width, cadence, walking speed, double-support time
• Fall risk prediction in elderly and neurological populations
• Cost-effective compared to full 3D motion analysis
• Sensor fusion (combining accelerometer + gyroscope data) improves accuracy

C. Wearable sEMG and Plantar Pressure Systems

• Surface EMG in wearable patches allows real-time muscle activation analysis during community ambulation
• Smart insoles with plantar pressure arrays quantify foot loading patterns and detect gait abnormalities
• Multi-modal wearable platforms combining IMUs + sEMG + plantar pressure sensors provide the most complete ambulatory gait picture

D. Machine Learning and Artificial Intelligence in Gait Analysis

• Deep learning (CNN, LSTM): Convolutional neural networks processing IMU time-series data can now estimate muscle activities during walking without laboratory EMG (Khant et al., Sci Rep, 2025), replacing invasive electrode placement
• Fall detection and prediction: TinyML algorithms optimised for microcontrollers enable edge-AI-based gait analysis for real-time fall risk alerts, particularly in elderly populations
• Pathological gait classification: ML algorithms trained on IMU and kinematic data can automatically classify gait disorders (Parkinson's, stroke hemiplegia, SCI patterns) with high accuracy
• Injury prediction in athletes: Multi-modal sensor data with ML accurately predicts running-related injuries by identifying critical gait features such as ground reaction force and stride length
• Automated gait event detection: AI algorithms identify gait cycle events (heel strike, toe-off) automatically, removing reliance on trained human observers

E. Robot-Assisted Gait Training (RAGT)

• Lokomat (Hocoma): Body-weight-supported treadmill with bilateral robotic exoskeleton; automates the stepping pattern while providing real-time kinematic feedback
• End-effector robots (e.g., Gait Trainer GT1, Morning Walk): Foot plates guide the foot through a physiological gait trajectory
• A 2024 meta-analysis (Chen et al., PMID: 38647534) showed RAGT significantly improves gait speed, balance, and kinematic parameters post-stroke
• A 2024 Cochrane-level systematic review (Mehrholz et al., PMID: 40365867) confirmed electromechanical-assisted training improves the odds of walking independently after stroke
• A 2024 RCT in JAMA Network Open (Choi et al., PMID: 39037815) demonstrated overground gait training with wearable robots improved walking in children with cerebral palsy

F. Body-Weight Supported Treadmill Training (BWSTT)

• Harness system suspended over a treadmill unloads a percentage of body weight
• Allows earlier gait training in patients who cannot support full weight
• Particularly beneficial in incomplete SCI, stroke, and TBI
• Can be combined with robotics (Lokomat) for complete gait cycle assistance
• Promotes neuroplasticity by providing repetitive, task-specific gait input

G. Virtual Reality (VR) and Augmented Reality (AR) in Gait Rehabilitation

• VR environments simulate real-world obstacles, uneven terrain, and dual-task challenges
• Provides visual biofeedback on gait parameters in real time
• Increases patient engagement and motivation (gamification)
• AR overlays gait feedback onto the patient's real-world view during overground training

H. Markerless Motion Capture

• Computer vision-based systems (e.g., OpenPose, MediaPipe) extract joint positions from standard video cameras without skin markers
• Enables large-scale gait screening in community settings (clinics, airports, elderly care homes) without expensive laboratory equipment
• AI-powered pose estimation algorithms have reached lab-comparable accuracy for many spatiotemporal parameters

I. Portable Gait Analysis for Neurological Conditions

• Parkinson's disease: IMU-based detection of freezing of gait (FoG), festination, and dual-task interference
• Multiple sclerosis: Remote monitoring of gait deterioration over time as a disease progression marker
• Stroke: Home-based gait monitoring during community reintegration phase

SUMMARY TABLE

REFERENCES

• Campbell's Operative Orthopaedics, 15th Ed. (2026) - Gait Analysis, CP Chapter
• JBLearning PT Manual: Wheelchairs, Assistive Devices, and Gait Training (O'Sullivan & Schmitz framework)
• Chen S et al. (2024). Robot-assisted gait training after stroke. Eur J Phys Rehabil Med. [PMID: 38647534]
• Mehrholz J et al. (2025). Electromechanical-assisted training for walking after stroke. Cochrane Database Syst Rev. [PMID: 40365867]
• Choi JY et al. (2024). Overground gait training with wearable robot in cerebral palsy. JAMA Netw Open. [PMID: 39037815]
• Bavan L et al. (2026). Wearable inertial sensors for paediatric gait assessment. Gait Posture. [PMID: 42105725]
• Gu B et al. (2026). Wearable sensor technologies for health monitoring. JMIR mHealth uHealth. [PMID: 41511829]

What's Inside (15 Slides)