The sensor — where every metric comes from

One capacitance grid under each cage

A board with a grid of electrodes (12×4 or similar) sits under each standard IVC, sensing the animal's position through dielectric (capacitance) change — no cameras, tethers, or handling. Sampling is sub-second, then aggregated into the metrics that follow. DVC-derived activity is comparable to video tracking [1], and the diurnal rhythm reproduces across three sites — CNR Rome, The Jackson Laboratory, Karolinska [2].

Iannello, Heliyon 2019 · Pernold et al., PLoS One 2019

Metric · Locomotion

Activation Density & Distance Walked

Iannello, Heliyon 2019 — foundational technology paper [1]

Metric · Activity budget & rest/activity bouts

Decomposing the day into bouts

Pernold et al., PLoS One 2023 — the baseline activity budget [3]

Metric · Rest Disturbance Index (RDI)

RDI is transdiagnostic

RDI quantifies the fragmentation and interruption of rest — derived from the same activity stream, without EEG/EMG. The same construct recurs across unrelated disease models:

Golini 2020 · Golini 2023 · Piilgaard 2023 · Moore 2024 [4–7]

Metric · Circadian / regularity

Day/night ratio, period, amplitude

Tir et al., Sci Rep 2025 — circadian phenotyping at rack scale [8]

Metric · Spatial preference

Frontality & wall activity are strain-specific

Beyond "how much," the grid records "where": time spent near the cage walls ("wall activity") and toward the front ("frontality"). These differ by strain — C57BL/6NCrl, BALB/c, and CD1 each show distinct spatial signatures — and were derived entirely from previously recorded data, with no new animals: a worked example of data repurposing for the Reduction principle [9].

Fuochi, Rigamonti et al., Sci Rep 2023 [9]

Environmental channel #1 · Bedding Status Index (BSI)

Bedding "wetness" from the same sensors

The electrode grid also senses soiled-bedding moisture. An ML model trained on human annotations predicts the Bedding Status Index ("wetness") at >90% accuracy at higher densities (5/cage). It supports data-driven cage-change intervals of 3–6 weeks (vs the standard 2-week schedule) and cut cage changes by 65–70% — with no effect on intracage ammonia, CO₂, growth, or circadian metrics [10].

Collins et al., JAALAS 2025 [10]

Environmental channel #2 · Urination Index

From bedding wetness to blood glucose

The same bedding-moisture signal becomes a metabolic biomarker. The Urination Index algorithm detects polyuria — a hallmark of diabetes — and correlates highly with blood glucose during hyperglycemia, enabling non-invasive, continuous, in-cage assessment of onset, progression, severity, and drug response, without stressful blood sampling [11].

Brachs et al., Lab Anim (NY) 2025 [11]

Optional metric · GYM500 wheel running

When bulk activity is not enough

An optional in-cage GYM500 running wheel logs running distance alongside DVC activity. In a cancer-induced bone pain model, wheel-running distance fell and tracked limb-use/weight-bearing scores, while general home-cage activity did not — making wheel running the more sensitive spontaneous pain-like readout, and showing exactly when to add the wheel [12].

Hopkins et al., In Vivo 2025 [12]

Synthesis · Reading the metric stack

Three patterns hold across the corpus

Pernold 2023 · Tomanelli 2024 · Collins 2025 · Brachs 2025 · Fuochi 2023 · Fuochi 2024 [3, 9–11, 13, 14]

References — full citations with DOIs

Sources