The human brain is a remarkably complex and flexible organ, with as many possible failure modes and glitches as there are emergent and surprising properties. Take something away, and sometimes you can coax another part of the brain to take up the slack in some other way. Case in point, artist Neil Harbisson. Harbisson was born with a condition called achromatopsia, which is the name for a group of disorders which collectively result in the same phenomenon - he cannot see colors, only shades of greyscale. Sometimes it's a neurological dysfunction, sometimes it's a defect in the retina, and sometimes it's due to some other combination of factors. In a decade-long experiment, Harbisson developed a prosthesis which incorporates a high resolution camera, a small radio scanner, and a microprocessor which converts patterns of color and RF emissions into patterns of sound. Over time, he's trained his brain to interpret those patterns of sound into a form that he can use to create his art. Some might say that he's trained himself to have a form of synaesthesia, or a state in which sensory modes are combined in unusual ways (such as seeing geometric or colored patterns when hearing certain sounds, or tasting phantom flavors when touching things with different textures). Recently, to increase the resolution of the new sensory input he had a new version implanted into his skull. The implant is much more sensitive, and its placement in his skull will give him much depth of sensation.
Of what use might this be to someone who isn't an artist? Ask someone who has a magnet implanted in their finger how useful it is when debugging power supplies. It's interesting to trace the evolution of brain implants. Sure, a lot of us would love to have datajacks, interfaces connected directly to our nervous systems that would hypothetically let us control communications devices and computers. There's some evidence that some aspects of such an interface are easier than previously thought, thanks to the nervous system's marvelous ability adapt to new forms of input, but there are other considerations that have to be taken into account. Chief among them are the fact that neurons that have been injured somehow (say, someone's poked a very tiny, very thin wire into them) will develop scarring and push the wires out. However, techniques to mitigate this are being worked on, and a significant amount of progress has been made toward interfacing with living neural networks. In testing right now, for example, are prosthetic versions of the hippocampus, the part of the mammalian brain which controls the storage and recall of long term memory. Deep brain stimulation units, which generate minute pulses of electricity and pump them through slender wires into the brains of patients are used to treat Parkinson's Disease, certain forms of obsessive-compulsive disorder, Alzheimer's Disease, and certain forms of clinical depression that neither therapy nor drugs have any effect upon.
There are other problems that we have to solve before cerebral implants are more common. For starters, the brain is exceptionally delicate for all of its power. Once you get the skull open and part the meninges the brain has a consistancy very similar to that of toothpaste. Think about that for a minute. Better yet, go to the bathroom and squeeze a little toothpaste onto your fingertip, and play with it a little. The cranial vault is not someplace one trespasses without extreme care. The blood-brain barrier protects the interior of the brain from the rest of the body; that's correct, I said the rest of the body. Biochemistry is an amazingly weird and responsive thing. The brain has a system of biochemistry all its own, and if the blood-brain barrier didn't isolate the brain from the rest of the body and enforce a strict set of exceptions and modulating factors, something as commonplace as a rise in blood glucose after dinner might wreak havoc with the brain's functioning. We don't yet have a complete model for how the blood-brain barrier operates, nor its functional parameters, so messing around inside the brain might have unforseen consequences that we haven't discovered yet. This means that, at a minimum, we have to be sure that the materials used in the construction of those implants are as bioneutral as we can make them. Power and heat dissipation are two other challenges that need to be met. Some implants in use today are externally powered, like cochlear implants. Others have self-contained power supplies which will require regular maintenance.
What's next? Good question. Sensory prostheses are an obvious direction to move in because medical science has been working on implants to restore vision to the blind and hearing to the deaf. Replacing the sense of touch is an ongoing project in the field of limb replacement because artificial limbs don't have a sense of touch, so sophisticated electrical engineering and signal processing is being done to figure out how best to convert mechanical pressure (say, from two digits holding a coin) into a signal that can be fed into the peripheral and central nervous systems. Replacing the senses of smell and taste aren't something we ordinarily think about but anosmia is a very real condition that causes people distress. Labs on a chip are a rapidly maturing technology that may see some application in the future for curing the condition. Medical science is already working on implementations of cortical stimulation and the prosthetic hippocampus shows that headway is being made reverse engineering certain functional structures of the brain. What else might be? Let's wonder out loud... what about being able to correct the condition known as blindsight, a condition that occasionally results from lesions in the striate (primary visual) cortex which mean that the brain is processing visual information from otherwise normal eyes, but the rest of the brain - the part that calls itself 'I' and 'me' - has no idea. Ask someone with blindsight and they'll probably tell you that they are without vision, and yet they interact with their environment as if they did. Splicing nerves back together after injury is another goal that we're having some success with, though the specific internal conditions necessary make it difficult.
The mind boggles at what may be.