Control in Bioengineering:
Applications to the Nervous System
COCHLEAR IMPLANTS
Outline
Figure 1: Illustration of a cochlear implant [4]
Speech is a complex of sound waves, consisting of multiple sound waves at differing frequencies and amplitudes. Together, they denote the pitch and volume of someone’s voice. The sound waves get transmitted through our ear to reach our cochlea, where the individual hair cells can pick out specific frequencies and vibrate at those resonant frequencies. The hair cells deeper in the cochlea will resonate at lower frequencies while those nearer to the entrance will resonate at higher frequencies, thus allowing our brain to determine the pitch of the sound from the location of these hair cells. [1] When resonance occurs, the hair cells excite the sensory nerves closest to them and the nervous impulse gets transmitted to the brain via action potentials.
When the hair cells get damaged or destroyed, they are no longer able to cause nervous impulses and this leads to deafness. In some cases, the sensory neurons close to these hair cells are still intact and cochlear implants can be used to replicate the function of the cochlea by exciting these neurons with artificial electrical impulses. [1]
Cochlear Implants as Open-Loop Control
Cochlear implants are essentially a form of open-loop control, where an input (sound) is converted to an output (electrical impulse). The implants always follow the same sequence from input to output and are unable to receive feedback about the quality of sound as registered by the patient’s brain. The patient has to work closely with medical personnel to ensure that the system is finely tuned so that they are able to understand speech. Fig 2 details the block diagram of a cochlear implant system.
The cochlear implant system consists of external and internal parts. The external part, which is worn around the ears, converts sound into electrical impulses.
Figure 2: Detailed block diagram of an open-loop cochlear implant system
-
A microphone detects sound and an analogue-to-digital convertor (ADC) converts sound waves into a digital signal that can be processed. [1]
-
The sound processor converts the sound waves from the time domain into the frequency domain, which results in a plot of the different amplitudes of sound at various frequencies. It will then use a speech processing strategy to reduce amplitudes at certain frequencies and amplify amplitudes at other frequencies. [2] This has the effect of amplifying frequencies that are most likely related to speech and reducing background noise. The sound processor is regularly tuned based on the patient’s feedback so that the specific speech processing strategy used is tailored to the individual patient. [3] The processor converts the result into a sequence of electrical impulses that will be used to stimulate nervous impulses in the cochlea.
-
The information for the impulse sequence is transmitted via radiofrequency waves into the internal implant.
-
A receiver receives the radiofrequency signal, decodes the information and sends the appropriate electrical current to the electrodes, which emits electrical impulses to generate a nervous response.
Development of Closed-Loop Cochlear Implant Systems
Closed-loop control, where the sound processor receives feedback and improves its speech processing strategies accordingly, would be beneficial to the patient, as they would be better able to understand conversations and would not need to have their processor re-tuned on a regular basis.
Currently, most cochlear implants have back telemetry [2], where a sensor in the electrode measures the actual electrical impulses and transmits this information via the same radiofrequency channel to the external unit. The processor then checks whether the output of the electrode matches the instructions that have been transmitted. Figure 3 shows a block diagram of a cochlear implant with back telemetry.
Figure 3: Block diagram of a cochlear implant system with back telemetry
This feedback loop monitors the quality of the nervous impulses generated in the auditory nerve and ensures that the voltage of the electrical impulses is not too high to harm the surrounding tissue. [2] However, back telemetry does not give information on the quality of the sounds heard, in other words, whether the sounds heard are intelligible as speech.
A recent improvement to the sensor allowed the detection of nervous impulses from the brainstem and the auditory cortex in addition to the auditory nerve. [3] With better sensors, sound processors would be better able to analyse the effectiveness of the implant. The next step would be to interpret the information received from the sensors and design an algorithm that would change the speech processing strategy automatically. Developments are currently underway and a closed-loop cochlear implant system might soon be possible.
References
1. Wilson, Blake S., and Michael F. Dorman. "Cochlear implants: a remarkable past and a brilliant future." Hearing research 242.1 (2008): 3-21.
2. Zeng, Fan-Gang, et al. "Cochlear implants: system design, integration, and evaluation." Biomedical Engineering, IEEE Reviews in 1 (2008): 115-142. URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4664429&isnumber=4689462
3. Laughlin, Myles Mc, et al. "Towards a closed-loop cochlear implant system: Application of embedded monitoring of peripheral and central neural activity."Neural Systems and Rehabilitation Engineering, IEEE Transactions on 20.4 (2012): 443-454.
4. Blausen.com staff. "Blausen gallery 2014". Wikiversity Journal of Medicine. DOI:10.15347/wjm/2014.010. ISSN 20018762. URL: https://upload.wikimedia.org/wikipedia/commons/5/50/Blausen_0244_CochlearImplant_01.png