diff --git a/img/pulseox_plans.jpg b/img/pulseox_plans.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..338adeaf985c917a19e3ed4ae4694f55e1c50ea7
Binary files /dev/null and b/img/pulseox_plans.jpg differ
diff --git a/pulseox/README.md b/pulseox/README.md
index 845abe7f3ab530cbae642759953f4cff3fb02021..b78874fb83805a53adf8e3395f7d104744155bee 100644
--- a/pulseox/README.md
+++ b/pulseox/README.md
@@ -29,6 +29,8 @@ See the Background section for more information on how these sensors work, the A
 - improve algorithm
     - filtering, curve fitting, etc
 
+![pulseox_plans](../img/pulseox_plans.jpg)
+
 ### Questions
 Is there a need for a crude, distributed-fab coronavirus early warning device? Something that, rather than giving an absolute SpO<sub>2</sub> reading, provides a go/no-go indication relative to an initial condition. The standard for commercial pulse-ox devices is master calibration in a clinical trial using human subjects that deliberately enter a hypoxic state. Instead, what about a cheap indicator that simply indicates the change in SpO<sub>2</sub> since its first measurement, using a green-yellow-red indication rather than a digital display? Such a device could be provided to patients to wear continuously upon hospital discharge (or an asymptomatic positive covid-19 test result), and would give them an indication when they need to call their doctor due to plunging blood oxygen content. Crucially, the device would be cheap and would not require specialized components, so it could be made easily anywhere.