**2. Audio-biofeedback**

Audio-biofeedback (ABFB) is a process through which the subject learns and regains the ability to control and influence their own physiological responses through one psycho-physiological feedback and greater proprioception [1]. Biofeedback is used in rehabilitation and is based on biomechanical and physiological measurements of the body such as the neuromuscular, respiratory, and cardiovascular systems, movements, postural control, and force. An example of physiological biofeedback is electromyography biofeedback to increase the activity in weak or paretic muscles or to reduce the tone in spastic muscles. In electromyography, surface electrodes are used to detect a change in skeletal muscle activity, which is then fed back to the user by a visual or auditory signal. Another example is cardiovascular biofeedback, which is used to reduce blood pressure in hypertension and lower the mean heart rate [2].

In ophthalmology, audio-biofeedback is used in low-vision rehabilitation of maculopathy [3]. In our study, patients received eight monocular training sessions of audio-biofeedback, each lasting 10 minutes, every 7 days. Microperimetry (MP-1, Nidek Tech., Padova, Italy) was performed at the beginning and at the end of the sequence, as well as ETDRS visual acuity (VA) at 4 meters and Pelli-Robson Contrast Sensitivity (CS) at 1 meter.

Audio-biofeedback employs a sound to train the patient to keep a specific gaze position, which is marked on the digital retinal image by the operator and displayed as a target to the patient. If the patient's gaze matches the selected position, a continuous sound is emitted. When the eye drifts away, the sound becomes progressively more discontinuous.

The contribution of audio-biofeedback to low-vision rehabilitation depends on the seriousness of the case, although it is useful in all cases. In some cases, where the PRL is in a good place, biofeedback allows stabilization of a fixation already used with the increase of retinal sensitivity in decibel and reduction of the fixation ellipse known as the bivariate contour ellipse area (BCEA).

In other cases, where the PRL used by the patient is in an area of little use for vision with an insufficient visual span to reading, audio-biofeedback allows shifting of the fixation to a better locus called the trained retinal locus (TRL).

#### **2.1 Features and parameters of audio biofeedback**

In audio-biofeedback with microperimetry, tracking is one of the key features, as it allows for automatic detecting of the patient's eye movements during the exam of fixation as well as during the feedback training. The tracking detects and scores the patient's fixation trajectory frame by frame. The user can preconfigure the parameters that define the characteristics of the fixation target, including the shape, extent, color, and thickness of the target.

There is the possibility to use letters or phrases as "custom" fixation.

The stability of fixation is classified [4–6] as follows:


The fixation itself is classified as follows:


The area of fixation is called bivariate contour ellipse area (BCEA) and is based on the published scientific literature [7]. The results of the relative analysis are converted into a graphical and numerical mode in three ellipses where the area and the measurements of each ellipse include different percentages of fixation points (68.2%, 95.4%, and 99.6%) corresponding respectively to 1–2-3 standard deviations. The BCEA is expressed in square degrees, the major and minor axes are expressed in degrees, and the inclination of the major axis is expressed from 90° to 90° with 0° for the horizontal position.

The normal value of BCEA is 0.5–1° squared.

In audio-biofeedback, the pattern used for rehabilitation training is a chessboard format with six alternating schemes with varying radius, frequency in Hertz (number of pattern image changes in each second), and degrees of retinal coverage from 2° to 8° with elements of 0.5° in size or more.

Patients are trained to fix the new area of the retina by asking them to move their gaze toward the new fixation. As the patient moves, an intermittent sound plays. The closer the patient gets to the new zone, the more the sound will be continuous.

#### **2.2 Case 1**

A 73-year-old woman presented with neovascular AMD treated with intravitreal anti-vascular endothelial growth factor therapy (anti-VEGF) in both eyes.

Best corrected visual acuity (BCVA) in the right eye was 20/400, and BCVA in the left eye was 20/500.

Microperimetry MP1 Nidek showed a spontaneous and unstable PRL localized below compared to atrophic fovea with a medium sensitivity of 8 dB in the right eye (**Figure 1**). The eccentricity of fixation measured by microperimetry was 4° (**Figure 2**). The PRL area represented by the BCEA was 111.28° squared (3 Std Dev) (**Figure 3**).

The left eye with a worse visual functioning situation showed a spontaneous PRL located below and nasally with a medium sensitivity of 7 dB (**Figure 4**).

**Figure 1.** *Microperimetry before rehabilitation.*

**Figure 2.** *Decentering of fixation before audio-biofeedback.*

The eccentricity of fixation was 6° (**Figure 5**) and BCEA was 71.89° squared (**Figure 6**).

The patient completed 10 sessions of audio-biofeedback lasting 10 minutes per eye. After the audio-biofeedback, there was an improvement in the quality of vision.

The best corrected visual acuity (BCVA) in the right eye improved from 20/400 to 20/200, and in the left eye it improved from 20/500 to 20/400. The contrast sensitivity and the parameters of microperimetry improved as well. In the right eye, the PRL shifted from below to the temporal position with an increase of mean sensitivity from 8 dB to 13 dB (**Figure 7**).

**Figure 3.** *BCEA before audio-biofeedback.*

**Figure 4.** *PRL and mean sensitivity.*

**Figure 5.** *Decentralization of fixation.*

**Figure 6.** *BCEA before audio-bofeedback.*

The BCEA (bivariate contour ellipse area) decreased from 111.28° squared (3 st SD) to 31.12° squared, thus showing a clear improvement of fixation stability that is closely related to the improvement in mean sensitivity and to the shift of the PRL (**Figure 8**).

Beyond improvement in visual acuity, contrast sensitivity, stability of fixation, and sensitivity, as well as relocated fixation, the patient acquired the awareness of his own ability to fix, to use the visual residual, and to move toward the best site of vision (**Figure 9**).

### **2.3 Case 2**

A 68-year-old woman presented with atrophic AMD. BCVA in both the right eye and left eye was 20/400.

The first microperimetry showed a central scotoma with a sensitivity of 0 dB without presence of PRL in the right eye but of an erratic searching of the presented fixation point without a precise point of fixation (**Figure 10**).

The BCEA area was 314.87° squared, and it was not possible to find the decentralization of fixation for the absence of fixation (**Figure 11**).

Microperimetry showed an unstable and spontaneous PRL located below the atrophic fovea with a mean sensitivity of 8 dB in the left eye (**Figure 12**).

The BCEA in the left eye was 90.10° squared (**Figure 13**).

The decentralization of fixation compared to the atrophic fovea was 7° (**Figure 14**).

The patient completed 10 sessions of audio-biofeedback in both eyes.

The purpose of the audio-biofeedback was to move the fixation up. In the right eye with erratic fixation, the cross was placed above the atrophic fovea, instead in the left eye the cross was placed above compared to the spontaneous PRL, which was located below the atrophic fovea (**Figure 15**). The passage from erratic fixation to superior fixation occurred gradually (**Figure 16**).

In the right eye, the fixation was shifted above and the sensitivity increased from 0 dB in the initial central scotoma with erratic fixation to 10 dB of mean sensitivity
