Sensors and Measurement Science

Validity is agreement with a criterion or gold-standard measure.

Source: Atkinson & Nevill, Sports Med 1998 · recall

A reliable device gives similar results when the true value has not changed.

Source: Hopkins, Sports Med 2000 · recall

A device with a fixed offset gives repeatable but biased readings.

Source: Atkinson & Nevill, Sports Med 1998 · recall

Bias shifts all readings up or down; random error scatters them around the true value.

Source: Atkinson & Nevill, Sports Med 1998 · recall

It plots the difference between two methods against their mean, showing bias and 95% limits of agreement.

Source: Bland & Altman, Lancet 1986 · recall

Typical error is the within-subject standard deviation across repeated trials.

Source: Hopkins, Sports Med 2000 · recall

Correlation measures association, not agreement; one device could read 20% higher throughout.

Source: Bland & Altman, Lancet 1986 · recall

Expressing error as a percentage lets reliability be compared across different measures.

Source: Hopkins, Sports Med 2000 · recall

If typical error exceeds the expected improvement, a test cannot detect it reliably.

Source: Hopkins, Sports Med 2000 · applied

Good reliability supports monitoring trends; the bias rules out direct comparison with the lab.

Source: Atkinson & Nevill, Sports Med 1998 · applied

Foot pods and watches estimate power from accelerometers, speed and body mass rather than measuring force.

Source: Cerezuela-Espejo et al., PLoS One 2020 · recall

Brands model running power differently, so their numbers are not interchangeable.

Source: Cerezuela-Espejo et al., PLoS One 2020 · recall

The Stryd foot pod showed the best repeatability, while agreement between technologies was poor.

Source: Cerezuela-Espejo et al., Eur J Sport Sci 2021 (five running power technologies) · recall

Power responds to gradient faster than heart rate, helping pace hills by effort.

Source: Cerezuela-Espejo et al., PLoS One 2020 · applied

Tendon recoil and changes in running form mean mechanical estimates relate loosely to energy used.

Source: Cerezuela-Espejo et al., PLoS One 2020 · recall

Only some systems include an air-resistance estimate, so headwind efforts may be under-reported.

Source: Cerezuela-Espejo et al., PLoS One 2020 · recall

Running and cycling watts are not equivalent, so each needs its own test.

Source: Cerezuela-Espejo et al., PLoS One 2020 · recall

Consistency within one device matters more than absolute values.

Source: Cerezuela-Espejo et al., PLoS One 2020 · applied

Mechanical power in running depends on the mass lifted and accelerated.

Source: Cerezuela-Espejo et al., PLoS One 2020 · recall

Algorithm changes can shift numbers, so trends across updates need care.

Source: Cerezuela-Espejo et al., PLoS One 2020 · applied

Casa found rectal (and gastrointestinal) measures valid; oral, aural, axillary and temporal were not.

Source: Casa et al., J Athl Train 2007 · recall

Mouth breathing, fluids and cooling of the ear canal cause large errors against rectal temperature.

Source: Casa et al., J Athl Train 2007 · recall

The capsule measures gastrointestinal temperature and transmits it wirelessly as it passes through.

Source: Byrne & Lim, Br J Sports Med 2007 · recall

Taken several hours before, the capsule reaches the intestine and is less affected by cold fluids.

Source: Byrne & Lim, Br J Sports Med 2007 · recall

Skin temperature responds to air, sweat and blood flow and does not track core reliably.

Source: Casa et al., J Athl Train 2007 · recall

Non-invasive devices infer core temperature from skin temperature and heat flow, with estimation error.

Source: Byrne & Lim, Br J Sports Med 2007 · recall

Oral or ear readings can look normal in a hyperthermic athlete, delaying cooling.

Source: Casa et al., J Athl Train 2007 · recall

Monitoring core temperature helps dose heat acclimation sessions safely and consistently.

Source: Byrne & Lim, Br J Sports Med 2007 · recall

Each pill is used once and can be affected by food, drink and passage time.

Source: Byrne & Lim, Br J Sports Med 2007 · recall

Estimates can be wrong; clinical signs and rectal measurement take precedence when heat illness is possible.

Source: Casa et al., J Athl Train 2007 · applied

CGMs sample interstitial fluid, not blood or muscle glycogen.

Source: Flockhart & Larsen, Sports Med 2024 · recall

Interstitial glucose trails blood glucose, especially when levels change quickly as in exercise.

Source: Flockhart & Larsen, Sports Med 2024 · recall

They were developed to help people with diabetes avoid dangerous highs and lows.

Source: Flockhart & Larsen, Sports Med 2024 · recall

The liver defends blood glucose, so readings can stay normal while muscle stores run down.

Source: Flockhart & Larsen, Sports Med 2024 · applied

Reviews find little evidence that CGM-guided fuelling improves performance in athletes without diabetes.

Source: Flockhart & Larsen, Sports Med 2024 · recall

Fast-changing glucose, altered blood flow and sweat can reduce sensor accuracy.

Source: Flockhart & Larsen, Sports Med 2024 · recall

Lying on the sensor and calibration limits can produce misleading low readings in healthy people.

Source: Flockhart & Larsen, Sports Med 2024 · recall

Normal glucose swings can be over-interpreted, leading to unnecessary eating or worry.

Source: Flockhart & Larsen, Sports Med 2024 · applied

Stress hormones prompt liver glucose release that can outpace uptake early in hard efforts.

Source: Flockhart & Larsen, Sports Med 2024 · recall

A single reading cannot show glycogen status; symptoms and the planned intake remain the guide.

Source: Flockhart & Larsen, Sports Med 2024 · applied

NIRS uses light absorption to estimate tissue oxygenation, often reported as SmO2.

Source: Perrey & Ferrari, Sports Med 2018 · recall

Both pigments absorb near-infrared light similarly, so signals combine.

Source: Perrey & Ferrari, Sports Med 2018 · recall

Light must pass through skin and fat; a thicker layer reduces the muscle contribution.

Source: Perrey & Ferrari, Sports Med 2018 · recall

Penetration depth is about half the distance between light source and detector.

Source: Perrey & Ferrari, Sports Med 2018 · recall

Different muscles and even parts of one muscle behave differently during exercise.

Source: Perrey & Ferrari, Sports Med 2018 · recall

Declining oxygen saturation suggests muscle demand exceeds local supply.

Source: Perrey & Ferrari, Sports Med 2018 · recall

NIRS breakpoints can coincide with other thresholds on average, but individual agreement varies.

Source: Perrey & Ferrari, Sports Med 2018 · recall

Different placement samples different tissue, making comparisons unreliable.

Source: Perrey & Ferrari, Sports Med 2018 · recall

Research uses NIRS to study oxygenation, blood flow and mitochondrial capacity non-invasively.

Source: Perrey & Ferrari, Sports Med 2018 · recall

NIRS-derived thresholds vary with placement and tissue, so they should be checked against other markers.

Source: Perrey & Ferrari, Sports Med 2018 · applied

A loss of 1 kg of body mass is about 1 litre of sweat, after adjusting for intake and output.

Source: Baker, Sports Med 2017 · recall

Sweat sodium varies several-fold between individuals, so population averages mislead.

Source: Baker, Sports Med 2017 · recall

Local sweat often has higher sodium than whole-body sweat; correction equations are used.

Source: Baker, Sports Med 2017 · recall

Acclimated athletes reabsorb more sodium in the sweat glands, lowering concentration.

Source: Baker, Sports Med 2017 · recall

Environmental conditions and effort can change sweat rate from under 0.5 to over 2 L per hour.

Source: Baker, Sports Med 2017 · recall

The sweat chloride test for cystic fibrosis reflects very salty sweat from impaired reabsorption.

Source: Baker, Sports Med 2017 · recall

Results from cool, low-intensity tests may not apply to a hot, hard race.

Source: Baker, Sports Med 2017 · applied

Visible salt suggests high sodium loss or high sweat volume, but only a test quantifies it.

Source: Baker, Sports Med 2017 · recall

Newer wearable patches with correction algorithms show reasonable agreement in studies but are still evolving.

Source: Baker, Sports Med 2017 · recall

Sweat loss = 1.8 + 0.6 = 2.4 L over 1.5 h, about 1.6 L per hour.

Source: Baker, Sports Med 2017 · applied

A capillary blood sample from the finger or earlobe is analysed on a test strip.

Source: Tanner, Fuller & Ross, Eur J Appl Physiol 2010 · recall

Devices agreed reasonably with lab analysers but differed from each other, so one device should be used consistently.

Source: Tanner, Fuller & Ross, Eur J Appl Physiol 2010 · recall

Sweat has high lactate concentrations, which can falsely raise readings.

Source: Tanner, Fuller & Ross, Eur J Appl Physiol 2010 · recall

Differences in circulation and sampling can create systematic gaps between sites.

Source: Tanner, Fuller & Ross, Eur J Appl Physiol 2010 · recall

Samples taken too early in a stage may not reflect the steady-state response.

Source: Tanner, Fuller & Ross, Eur J Appl Physiol 2010 · recall

Milking the finger pushes interstitial fluid into the drop, lowering and distorting the reading.

Source: Tanner, Fuller & Ross, Eur J Appl Physiol 2010 · recall

Only consistent equipment and procedures make changes between tests meaningful.

Source: Tanner, Fuller & Ross, Eur J Appl Physiol 2010 · recall

Low glycogen, prior exercise and stage length change lactate at a given intensity.

Source: Tanner, Fuller & Ross, Eur J Appl Physiol 2010 · recall

Onset of blood lactate accumulation at 4 mmol/L is a convenient reference that varies in meaning between athletes.

Source: Tanner, Fuller & Ross, Eur J Appl Physiol 2010 · recall

Glycogen status changes lactate production, so pre-test diet should be standardised.

Source: Tanner, Fuller & Ross, Eur J Appl Physiol 2010 · applied

Chest straps detect the ECG signal through electrodes on the skin.

Source: Gillinov et al., Med Sci Sports Exerc 2017 · recall

Photoplethysmography uses light absorption to detect pulsatile blood flow.

Source: Gillinov et al., Med Sci Sports Exerc 2017 · recall

The chest strap agreed best with ECG; wrist devices varied, especially on the elliptical and arms-involved activities.

Source: Gillinov et al., Med Sci Sports Exerc 2017 · recall

Rhythmic motion can be mistaken for the pulse, so the sensor reports cadence.

Source: Gillinov et al., Med Sci Sports Exerc 2017 · recall

Vasoconstriction in the cold lowers peripheral blood flow at the wrist.

Source: Gillinov et al., Med Sci Sports Exerc 2017 · recall

Movement and water between sensor and skin reduce accuracy in the pool.

Source: Gillinov et al., Med Sci Sports Exerc 2017 · recall

Less movement and better contact on the upper arm reduce motion artefacts.

Source: Gillinov et al., Med Sci Sports Exerc 2017 · recall

Ink in the skin absorbs light and can interfere with the optical signal.

Source: Gillinov et al., Med Sci Sports Exerc 2017 · recall

Dry electrodes and static from clothing can produce spurious readings early on.

Source: Gillinov et al., Med Sci Sports Exerc 2017 · recall

Readings near step rate during easy running are a classic sign of cadence lock.

Source: Gillinov et al., Med Sci Sports Exerc 2017 · applied

Recording points at intervals and smoothing the track cuts off corners, shortening measured distance.

Source: Ranacher et al., Int J Geogr Inf Sci 2016 (GPS distance measurement error) · recall

Dual-frequency (multi-band) reception helps reject reflected signals in difficult environments.

Source: Bowditch, American Practical Navigator (2019) · recall

Multipath reflections lengthen the apparent signal path, causing position errors.

Source: Bowditch, American Practical Navigator (2019) · recall

Rotations times wheel circumference give distance without satellite errors, if circumference is set correctly.

Source: Bowditch, American Practical Navigator (2019) · recall

GPS vertical accuracy is poorer than horizontal; pressure sensors track relative height well but drift with weather.

Source: Bowditch, American Practical Navigator (2019) · recall

A passing weather front changes pressure and appears as a change in height unless recalibrated.

Source: Bowditch, American Practical Navigator (2019) · recall

Inertial sensors estimate stride and speed; calibration on a known distance improves accuracy.

Source: Cerezuela-Espejo et al., PLoS One 2020 · recall

Inertial sensors work where GPS cannot reach.

Source: Cerezuela-Espejo et al., PLoS One 2020 · recall

An error of 1% in circumference gives a 1% error in all speed and distance data.

Source: Bowditch, American Practical Navigator (2019) · recall

Certified courses are measured on the shortest line; watch error and running wide add distance.

Source: Bowditch, American Practical Navigator (2019) · applied

An unusually poor test is likely to be followed by a better one by chance alone, which can mimic a treatment effect.

Source: Barnett et al., Int J Epidemiol 2005 · recall

Knowing which treatment is received can create placebo or nocebo effects.

Source: Hopkins et al., Med Sci Sports Exerc 2009 · recall

Participants act as their own controls, which increases sensitivity with small samples.

Source: Hopkins et al., Med Sci Sports Exerc 2009 · recall

Without washout, carry-over from the first treatment can distort the second.

Source: Hopkins et al., Med Sci Sports Exerc 2009 · recall

Randomising conditions repeatedly in one person tests what works for that individual.

Source: Hopkins et al., Med Sci Sports Exerc 2009 · recall

A tiny effect can be statistically significant in a large sample yet irrelevant to athletes.

Source: Hopkins et al., Med Sci Sports Exerc 2009 · recall

Studies showing effects appear more often in print, which inflates perceived benefits.

Source: Hopkins et al., Med Sci Sports Exerc 2009 · recall

Multiple comparisons raise the chance of false positives unless corrected.

Source: Hopkins et al., Med Sci Sports Exerc 2009 · recall

A bad race is often followed by a better one anyway, and belief can improve performance.

Source: Barnett et al., Int J Epidemiol 2005 · applied

Findings depend on the population, sport and conditions studied.

Source: Hopkins et al., Med Sci Sports Exerc 2009 · applied