TFT displays require continuous refreshing of their content. The voltage for each pixel is stored in a small capacitor on the display; this discharges over time. If a memory cell is placed in place of the capacitor, the information can be retained until the display is de-energized. This technology is known as "MIP" - Memory In Pixel.
The very low power consumption - only 1/50 even compared to an STN display - makes the technology especially suitable for battery-powered devices. These can be "wearables", but also measuring devices that need to be in operation for several years independently of the mains.
... are available in monochrome versions and as color displays with 64 colors. The diagonals range from just over 1" up to 4.4" and resolutions up to a maximum of 280x280. The displays are generally reflective or weakly transmissive, i.e. without backlight, which is the largest energy consumer in a display system.
The optical properties are adapted for use in ambient light; with a slightly transmissive polarizing filter, they can be read in low incident light with a low power consumption backlight.
Compared to other display technologies, the MIP performs well in terms of power. Assuming similar displays, an STN module without backlight requires 190mW and the MIP 1.2mW - a factor of 150. The logic alone of the TFT requires three times the power; it cannot be operated without backlight due to its transmissive properties. The STN display brings up the rear in terms of optical properties.
The interface is simple enough to drive it directly with a microcontroller. The memory function means that the content only needs to be updated when changes are made.