ORPHE INSOLE sample data

Walking and running sample data

Normal walking, 8 simulated pathological gaits and treadmill running at 3 speeds. See the acceleration, angular velocity and plantar pressure of both insoles on the same clock as the video. CSV, ARKit skeleton and video can be downloaded all at once or per trial.

Please read the terms of use before downloading. The data may not be redistributed.

12 trials Normal walking, 8 simulated pathological gaits, running at 3 speeds
208.3 Hz Acceleration, angular velocity, quaternion and 6-channel pressure, both feet
ARKit 30 fps 91-joint skeleton and video (faces mosaicked)
Synchronized A jump at the start aligns both feet and the video

Trials

12 trials, in video and waveforms.

Walking trials are about 5 m straight toward the camera; running is on a treadmill. Each trial starts with a two-footed jump. The pathological gaits are simulated by one healthy adult, and unilateral gaits use the left foot as the affected side.

Normal walking

The baseline for comparing the other trials.

Bilateral gaits (simulated)

Both feet use the same movement. Short-stepping and freezing are gait disorders seen in conditions such as Parkinson's disease.

Unilateral gaits (simulated, left = affected)

Gaits with related causes sit side by side. Circumduction and steppage gaits are swing impairments from hemiplegia or foot drop; Trendelenburg and Duchenne gaits come from weak hip abductors.

Treadmill running

6 / 9 / 12 km/h. Stride time stays almost constant while stride length, impact and push-off grow.

02 · Bilateral (simulated) · 21.3 s

Short-stepping gait

Strides about a quarter of normal length. The heel does not lift at push-off — close to shuffling.

03 · Bilateral (simulated) · 13.3 s

Freezing of gait

About 4 s of freezing at gait initiation, stepping in place before walking off.

05 · Unilateral (simulated, left) · 19.1 s

Circumduction gait

The affected leg swings outward in an arc. Swing width 26 cm.

06 · Unilateral (simulated, left) · 20.2 s

Steppage gait (foot drop)

The toes stay dropped and the forefoot lands first (strike angle +18°).

08 · Unilateral (simulated, left) · 20.2 s

Trendelenburg gait

The affected foot is supported on its lateral edge. Little left–right timing difference.

09 · Unilateral (simulated, left) · 20.7 s

Duchenne gait

Unlike the other unilateral gaits, stance is longer on the affected side.

Viewer

Video and waveforms on one clock.

Play, seek or click the charts and the video and waveforms move together. t = 0 is the left-foot jump landing; the right foot and the video are corrected with offsets found from the jump.

ZIP for this trial
t = 0.00 s

Space: play / pause · ← →: step one frame · click or drag a chart to seek

Normal walking (5 m)

Waveforms

  • gyroX: negative right after contact (the sole rolls onto the floor) → near 0 in mid-stance → a large negative peak just before toe-off (push-off) → a positive peak in mid-swing (toes lift)
  • Pressure: load moves from the heel (6) to the lateral midfoot (5) and the lateral toe (3). The total has two peaks: heel strike and push-off
  • Acceleration: an accZ peak at contact; accY turns negative in late swing (slowing before contact)

What the app metrics show

  • Stride length 1.3 m, speed 1.1–1.2 m/s → a normal walking speed, above the frailty threshold (1.0 m/s)
  • Stance 64% → close to the typical 60%
  • Strike angle −15 to −18°, toe-off angle −68° → lands on the heel and lifts it well at push-off
  • Foot height 15 cm → ample clearance

Short-stepping gait

Waveforms

  • Small gyroX amplitude; the push-off and swing peaks are gentle
  • accY swings strongly negative at contact (a shuffling contact)
  • Pressure is low overall and the heel carries almost no load. Both feet move alike, with little left–right difference

What the app metrics show

  • Stride length 0.31–0.32 m (about 1/4 of normal), speed about 0.4 m/s → many steps but little forward progress
  • Stride time 0.75–0.79 s, swing time 0.24 s (normal 0.39 s) → each swing is short and quick
  • Foot height 6 cm (normal 15 cm), toe-off angle −27 to −29° (normal −68°) → the heel does not lift at push-off, close to shuffling
  • Strike angle 0° → lands with the whole sole rather than the heel

Freezing of gait

Waveforms

  • For about 4 s right after trying to start (t ≈ 4.8–8.6 s, shaded in the charts) the subject cannot move forward and steps in place. Angular velocity only shows small ±50 dps oscillations, with no push-off or swing waveform
  • Both feet stay loaded during the freeze, mainly on the lateral toe (3) and the lateral midfoot (5). The heel (6) carries almost nothing — the weight is on the forefoot
  • Once the freeze ends, a few large gyroX waveforms appear and accY swings strongly negative at each contact

What the app metrics show

  • The first stride, which includes the freeze, is recorded as stride time about 2.4 s, stance 88% and speed 0.2 m/s → the app shows it as one extremely long, slow stride
  • After the freeze: stride time 0.85–0.87 s, stride length 0.45–0.72 m, strike angle 0 to +1°, toe-off angle −29 to −47° → no heel strike; shorter and shallower than normal
  • Foot height 5–10 cm (normal 15 cm), contact acceleration 19–20 m/s² (normal 26–32) → the feet stay low and land softly
  • Only 1–2 strides per foot can be analyzed, so treat the values as reference only

Scissor gait

Waveforms

  • Both feet load mainly on the lateral midfoot (5), with little load on the heel
  • Swing-phase angular velocity varies from stride to stride

What the app metrics show

  • Swing width: left 17 cm, right 9 cm (normal 2 cm) → both feet deviate sideways during swing
  • Foot angle: +28° / +13° (normal −14 to −16°) → foot orientation changes a lot on both feet
  • Strike angle +9 to +11° → both feet land toe-first
  • Stance 75–78% (normal 64%) → long support on both feet, a cautious gait
  • Contact acceleration: left 68 m/s², right 29 m/s² → despite the same intended movement, the left foot lands harder

Circumduction gait

Waveforms

  • Affected side: gyroX is almost flat during swing — the leg swings without moving the ankle
  • Affected side: supported by the heel (6) alone, with no load transfer to the forefoot (no push-off)

What the app metrics show

  • Swing width: affected 26 cm (unaffected 1 cm, normal 2 cm) → the leg swings outward. The metric that best captures circumduction
  • Stance: affected 44%, unaffected 73%; swing time: affected 0.86 s (unaffected 0.42 s) → short support and a slow swing on the affected side, so the unaffected side supports longer
  • Toe-off angle: affected −13° (unaffected −30°); push-off angular velocity: affected −194 dps (unaffected −357 dps) → no ankle push-off; the whole leg is lifted off
  • Contact acceleration: affected 94 m/s² (unaffected 45, normal 26–32) → the foot slams down

Steppage gait (foot drop)

Waveforms

  • Affected side: the lateral toe (3) responds first, then the load moves to the heel (6)
  • Affected side: gyroX is positive right after contact (the heel drops after the forefoot lands); almost flat during swing

What the app metrics show

  • Strike angle: affected +18° (normal −15 to −18°) → the forefoot lands first with the toes dropped. The metric that best captures steppage gait from foot drop
  • Stance: affected 50%, unaffected 78%; swing time: affected 0.71 s (unaffected 0.32 s) → a slow swing on the affected side and long support on the unaffected side
  • Toe-off angle: affected −22° (unaffected −52°); push-off angular velocity: affected −183 dps (unaffected −390 dps) → weak push-off on the affected side
  • Swing velocity: affected 1.3 m/s (unaffected 2.4 m/s) → the affected leg swings slowly

Antalgic gait

Waveforms

  • Short stance on the affected side (less time bearing weight)
  • Small gyroX amplitude and a weak push-off on the affected side

What the app metrics show

  • Stance: affected 60%, unaffected 71%; swing time: affected 0.58 s (unaffected 0.40 s) → less time with weight on the painful side
  • Push-off angular velocity: affected −162 dps (unaffected −387 dps) → no strong push-off on the affected side
  • Foot height: affected 5 cm (unaffected 9 cm); swing velocity: affected 1.2 m/s (unaffected 1.8 m/s) → the affected leg swings low and slowly

Trendelenburg gait

Waveforms

  • Affected side: supported mainly by the lateral midfoot (5) and the lateral toe (3); the heel carries almost nothing (lateral-edge support)
  • Affected side: accX shows a gravity offset during stance (the foot tilts outward)

What the app metrics show

  • Timing is nearly symmetric (stance 66 / 65%, swing 0.55 / 0.54 s) → little left–right timing difference; the actual abnormality is pelvic and trunk tilt
  • Contact acceleration: affected 18 m/s² (unaffected 31) → the affected foot is set down softly
  • Toe-off angle: affected −46° (unaffected −36°) → push-off is preserved
  • App metrics alone show little; read them with the lateral plantar loading. Pelvic tilt, the main sign, hardly appears in foot sensors

Duchenne gait

Waveforms

  • Opposite to the other unilateral gaits, stance is longer on the affected side
  • gyroX during swing is very small on the unaffected side

What the app metrics show

  • Stance: affected 72%, unaffected 54%; swing time: unaffected 0.76 s (affected 0.46 s) → long support on the affected side while the unaffected leg swings slowly
  • Unaffected swing velocity 1.2 m/s (affected 2.3 m/s), push-off angular velocity −213 dps (affected −294 dps) → the unaffected swing is small and slow
  • Foot angle: unaffected −22° (affected −11°) → the unaffected foot changes orientation
  • Lateral trunk lean, the main sign, hardly appears in foot sensors

Treadmill running 6 km/h

Waveforms

  • As speed increases, contact time shortens and the peaks of gyroX, acceleration and total pressure grow
  • Contact starts at the heel (6) and moves quickly to the forefoot. The faster the run, the more load on the forefoot (1, 3)
  • Unlike walking, sensors 1, 2 and 4 also respond

What the app metrics show

  • From 6 → 9 → 12 km/h, stride length goes 1.1 → 1.6 → 2.1 m while stride time stays nearly constant at 0.73–0.75 s
  • Stance 37 → 34 → 30%, toe-off angle −43 to −50 → −57 to −62°, foot height 21–25 → 46–51 cm
  • Contact acceleration 70–78 → 83–115 m/s², push-off angular velocity −426 to −430 → −752 to −811 dps, swing velocity 3.5–3.7 → 5.9–6.0 m/s
  • On the treadmill, stride length and speed read 9–16% below the belt speed (1.67 / 2.50 / 3.33 m/s)

Treadmill running 9 km/h

Waveforms

  • As speed increases, contact time shortens and the peaks of gyroX, acceleration and total pressure grow
  • Contact starts at the heel (6) and moves quickly to the forefoot. The faster the run, the more load on the forefoot (1, 3)
  • Unlike walking, sensors 1, 2 and 4 also respond

What the app metrics show

  • From 6 → 9 → 12 km/h, stride length goes 1.1 → 1.6 → 2.1 m while stride time stays nearly constant at 0.73–0.75 s
  • Stance 37 → 34 → 30%, toe-off angle −43 to −50 → −57 to −62°, foot height 21–25 → 46–51 cm
  • Contact acceleration 70–78 → 83–115 m/s², push-off angular velocity −426 to −430 → −752 to −811 dps, swing velocity 3.5–3.7 → 5.9–6.0 m/s
  • On the treadmill, stride length and speed read 9–16% below the belt speed (1.67 / 2.50 / 3.33 m/s)

Treadmill running 12 km/h

Waveforms

  • As speed increases, contact time shortens and the peaks of gyroX, acceleration and total pressure grow
  • Contact starts at the heel (6) and moves quickly to the forefoot. The faster the run, the more load on the forefoot (1, 3)
  • Unlike walking, sensors 1, 2 and 4 also respond

What the app metrics show

  • From 6 → 9 → 12 km/h, stride length goes 1.1 → 1.6 → 2.1 m while stride time stays nearly constant at 0.73–0.75 s
  • Stance 37 → 34 → 30%, toe-off angle −43 to −50 → −57 to −62°, foot height 21–25 → 46–51 cm
  • Contact acceleration 70–78 → 83–115 m/s², push-off angular velocity −426 to −430 → −752 to −811 dps, swing velocity 3.5–3.7 → 5.9–6.0 m/s
  • On the treadmill, stride length and speed read 9–16% below the belt speed (1.67 / 2.50 / 3.33 m/s)

Dataset

What is inside, and how the clocks are aligned.

Each ZIP has one folder per trial with sensor values, the app's gait analysis, the ARKit skeleton and the video. The same description is included as README.md.

Recorded
21 October 2026
Device
ORPHE INSOLE β (left and right). Acceleration [G], angular velocity [dps], quaternion and 6-channel pressure [N] at 208.3 Hz
Camera
iPhone ARKit body tracking (about 30 fps, 91 joints) and video. Faces are mosaicked
Subject
One healthy adult simulating each gait. Unilateral gaits simulate the left foot as the affected side; short-stepping, freezing and scissor gaits use the same movement on both feet
Trials
Walking is about 5 m straight toward the camera; running is on a treadmill. Each trial starts with a two-footed jump for synchronization
File Content
sensor-left.csv
sensor-right.csv
Sensor values. timestamp [s, UNIX time], accX/Y/Z [G], gyroX/Y/Z [dps], quatW/X/Y/Z, press1–press6 [N]
gait_analysis_left_detailed.csv
gait_analysis_right_detailed.csv
Per-stride gait metrics computed by the app
landmarks.json ARKit skeleton (3D joint positions from the camera origin [m], about 30 fps)
video.mp4
poster.jpg
Video and thumbnail (faces mosaicked)
trial.json
trials.csv
Trial name and synchronization values. trials.csv puts all trials in one table (full archive only)
Coordinate system ORPHE INSOLE with the sensor coordinate frame drawn on it: Y toward the toes, X to the right of the toe direction, Z upward out of the insole surface.

Right-handed and Z-up, the same for both feet. Y points toward the toes, X to the right of the toe direction, Z up (accZ ≈ +1 G at rest). +X is medial on the left foot and lateral on the right foot, so flip the sign of accX, gyroY and gyroZ on the left foot to align the feet anatomically.

Pressure sensors Top view of left and right ORPHE INSOLE with pressure sensors numbered 1 to 6.

1 toe (medial), 2 ball (medial), 3 toe (lateral), 4 midfoot (center), 5 midfoot (lateral), 6 heel. Values below about 9 N read as 0, and the total during quiet standing is only 10–20 N. Read it as load distribution and timing, not absolute load. Sensor 4 barely responds in this data.

Synchronization

Align both feet and the video with the opening jump

The insoles and the video run on separate clocks. The right foot is corrected by cross-correlating the jump landing impact (offsets 19–276 ms, about ±10 ms), and the video by the difference to the landing found from the vertical motion of the ARKit pelvis joint (0.27–0.36 s, about ±1 frame).

t_left  = timestamp − tZeroUnix
t_right = timestamp − rightOffsetMs / 1000 − tZeroUnix
video   = t + (tZeroUnix − videoStartUnix) + videoOffsetS

App gait metrics

Compare the trials by their metrics.

Medians over the steady part of gait_analysis_*_detailed.csv (left / right). Click a row to open that trial in the viewer.

Freezing of gait uses only the 1–2 strides after the freeze ended (reference only). A negative strike angle is a heel strike and a positive one a toe strike. The event-based columns are initialContactAccelerationMax.z, toeOffAngularVelocityMin.x and swingPhaseStrideVelocityMax.y.

Temporal–spatial
  • Stride time: time from one contact to the next contact of the same foot — the walking rhythm
  • Stance / swing: the share of a stride on and off the ground. Typical walking is 60% stance and 40% swing (20% double support), and double support is reported to rise to about 30% with frailty [Schwenk M, 2014]. A short stance and long swing on the affected side means the subject avoids bearing weight on it
  • Stride length: the distance between contacts of the same foot (about 1.3 m in normal walking) — about twice the alternating step (about 70 cm on average [Götz-Neumann K, 2005])
  • Speed: a usual gait speed below 1.0 m/s is one diagnostic criterion for frailty and sarcopenia [Satake S & Arai H, 2020], and below 0.7 m/s is associated with a higher fall risk [Shimada H, 2009]
Foot motion
  • Strike angle: angle between the sole and the floor at contact. Negative is a heel strike (normally −15 to −18°), near 0 the whole sole, positive a toe strike
  • Toe-off angle: the angle when the foot leaves the floor. The larger its magnitude, the more the heel lifts at push-off (normally −68°)
  • Max foot height: how high the insole rises during swing — a clearance measure; small values mean shuffling (normally 15 cm)
  • Swing width: sideways deviation during swing (positive = outward). It grows with circumduction and scissor gaits
  • Foot angle: toe direction relative to the walking direction. Its sign convention is still being confirmed, so compare it across conditions or feet
Event-based
  • Contact acceleration: the maximum Z acceleration around contact. Larger means a harder landing (normally 26–32 m/s²)
  • Push-off angular velocity: the minimum gyroX around toe-off (negative = toes rotating down). The larger its magnitude, the stronger the push-off (normally −350 to −505 dps)
  • Swing velocity: the maximum forward foot velocity during swing — how fast the leg swings (normally 2.9–3.9 m/s)

Terms of use

Please read before use.

The data is copyright ORPHE Inc. The same terms are included in README.md in each ZIP.

  1. Permitted use: free for education and workshops, academic and non-commercial research, personal creative projects, and prototypes and internal experiments
  2. Commercial use: to use the data to develop or evaluate a paid product or service, contact ORPHE beforehand
  3. No redistribution: do not redistribute the data itself (CSV, JSON, video, images). To share it, link to this page. Figures, statistics and analysis results derived from the data may be used in papers and presentations
  4. Prohibited: attempting to identify the subject, removing the mosaic or recovering the original video, and using the video or images to identify people or to train generative models
  5. Attribution: when you use the data in a paper, presentation or publication, cite “ORPHE INSOLE sample data (ORPHE Inc., 2026)” with the URL of this page
  6. Not for medical use: this is simulated reference data, not clinical reference values. ORPHE INSOLE is not a medical device and must not be used as a basis for diagnosis or treatment
  7. No warranty; changes: the data is provided “as is”, without warranty of accuracy. Its content may change, or publication may stop, without notice

Please note

Simulated by one healthy adult.

These are not clinically representative values and cannot be used for diagnosis. landingForce in the app output is 0 on every row (not computed), and the sign convention of footAngle is still being confirmed.

Send feedback