The following is the first part of a series on brain-machine integration and biomechanical solutions to restore function to tissues damaged by disease, trauma, or time.

Recent advances in prosthetic design for upper-limb amputees may restore thermal sensation in the near future. Over 180,000 people undergo limb amputation in the United States every year. An estimate from 2017 claims that over 57 million people worldwide had undergone limb amputation due to traumatic causes, a number that has likely risen due to global population growth.

Wearable prosthetic limbs, while useful, are often cumbersome and lack the touch sensation human limbs enable. Amputees are typically subjected to loss of touch sensation for the remainder of their lives. However, this may no longer need to be the case.

Researchers Iberite et al. developed a noninvasive wearable device to restore thermal sensation to upper-limb amputees. The combination of phantom thermal sensation and haptic feedback artificial skin would be a major breakthrough in prosthetic applications. Here we discuss the development of Iberite et al.’s thermal sensation device.

Presence of thermal phantom hand maps in upper-limb amputees

A commonly reported sensation by amputees is that of phantom limb sensation. When touched by a pen on their residual arm at different pressure points, some may experience the feeling of touch where their fingers once were. Of 27 participants, Iberite et al. found that 26 reported phantom sensations when prompted.

Phantom thermal sensation is also experienced by amputees. Using a custom thermal device, the researchers developed thermal maps of where phantom thermal sensation was detected in the 27 participants. Some only felt hot or cold sensations, whereas some felt no thermal sensation at all. They developed maps for 15 of 27 participants, which could be implemented into prosthetic hands.

Thermal phantom sensations are phenomenologically similar to those of the intact hand

To test the consistency of their thermal maps, Iberite et al. compared them to the real thermal sensation of the mirrored nonamputated arm of the participants. The phantom thermal sensations on the amputated side should match the real sensation on the nonamputated side.

The researchers began with nonamputee control subjects to ensure the thermal wearable was delivering consistent thermal sensations. The participants reported 76.4% of stimuli to match between each side of their body, most closely matching were sensations delivered by plastic (86.8%), followed by glass (72.9%) and copper (68.75%) as compared to the wearable. This means real sensation and phantom sensation felt similar to nonamputees just over three-quarters of the time.

These results were successfully replicated with the amputees, detecting similar sensations 83.3% of the time. The highest rate of similar sensation in the amputee cohort was copper (90%), followed by plastic and glass (both at 80%). This finding suggests phantom sensations of amputated limbs and real sensations of nonamputated limbs in the same individual are phenomenologically similar.

Delivering thermal phantom sensations through a wearable device

The phantom thermal maps were applied to a wearable sensor dubbed MiniTouch. Applying MiniTouch to nine participants, Iberite et al. developed two tasks.

The first task was to identify the differing temperatures of three glasses of water (15°, 24°, or 40°C) as one of four options (cold, cool, neutral, and warm/hot) using MiniTouch. The nine participants, on average, correctly identified 97.2% of the warm/hot (40°C) and cold (15°C) stimuli and were exactly correct 71.3% overall, on average.

The second task was to identify the material (copper, glass, or plastic) touching the sensor of the MiniTouch. Participants were successful 65.93% of the time, well above the statistically significant threshold of 40%.

Thermal phantom sensations are partially stable over months

Most encouragingly, phantom thermal sensations remained consistent over several months of recurring visits. Conducting stability testing for two of the participants two and four months after the development of their thermal maps, the phantom sensation was at least partially stable for both, suggesting that a wearable thermal device could be continually used for longer durations, only requiring minor adjustments a few times a year.


In a previous entry in this series, we discussed the recent advances in artificial skin technology and the ability to relay haptic feedback to the motor cortexes of the brain for rapid sensation. Wearable thermal devices are due similar praise in the field of regenerative medicine. Combining such technologies would enable amputees to regain significant touch sensation once thought lost. Beyond prosthetic applications, thermal sensation could be applied to robotics, manufacturing machinery, and other various commercial and medical applications in short order. We eagerly await the further development of this technology.