Technique for determining a swallowing deficiency

1. A method for training a processing unit for assessing characteristics of a swallowing process in a subject, the method comprising:

providing a kit comprising at least one device comprising the processing unit and a sensor capable of detecting throat vibrations externally or internally on a throat of the subject, the kit further comprising a thickening agent formulated for thickening a fluid for producing at least one item of foodstuff with a defined texture for intake by the subject, the processing unit after the training is capable of outputting an indication of the characteristics of the swallowing process in the subject,

wherein the processing unit is configured to accept a sensor signal associated with an intake of an item of tagged foodstuff by the subject, wherein the processing unit accepts the sensor signal by a process comprising the steps of:

a) performing at least once the following steps:

i) providing to the subject at least three items of tagged foodstuff for oral intake, each of the at least three items of tagged foodstuff selected from the group consisting of water, a foodstuff item with a texture having a nectar-type viscosity of 51 mPa s to 350 mPa s, a foodstuff item having a honey-type viscosity of 301 mPa s to 1750 mPa s, and a foodstuff item having a pudding-type viscosity of more than 1750 mPa s,

ii) assessing the characteristics of the swallowing process in the subject resulting from the intake of the at least three items of tagged foodstuff by the subject with video fluoroscopy; and

iii) using the results of step i) and ii) for training the processing unit, the training comprising determining the sensor signal resulting from the throat vibrations resulting from the intake of at least two items of tagged foodstuff by the subject, and forwarding the sensor signal to the processing unit, the processing unit is operated by a computer executable software, and the computer executable software provides an assessment of the characteristics of the swallowing process, and the assessment from the computer executable software results from correlating the sensor signal with the assessment resulting from the video fluoroscopy,

repeating steps i) to iii) at least once to improve correlation between an assessment of the characteristics of the swallowing process of the subject resulting from the sensor signal and the assessment of the characteristics of the swallowing process resulting from the video fluoroscopy, wherein the subject is selected from the group consisting of the same subject and a different subject; and

b) thereby obtaining a trained processing unit,

wherein the method comprises at least one step selected from the group consisting of (A) implementing a neural network and (B) determining parameters of the computer executable software running on and instructing the trained processing unit.

2. The method of claim 1 , wherein step iii) comprises:

a) providing the trained processing unit;

b) performing the method defined in step a) on the trained processing unit on a group of 10 to 30 subjects;

c) thereby obtaining a further trained processing unit; and

d) resubmitting the further trained processing unit to steps b)-d) for 1-30 times.

3. The method of claim 1 , further comprising a step c) wherein the trained processing unit is validated in a validation phase.

4. The method of claim 3 , comprising a further step of validating that the trained processing unit provides the same assessment of the characteristics of the swallowing process of the subject as the video fluoroscopy by comparing the indication of a swallowing deficiency output by the video fluoroscopy and output by the trained processing unit,

and thereby obtaining a trained and validated processing unit.

5. The method of claim 1 , wherein the trained processing unit is configured for at least one of (i) outputting an indication of a swallowing deficiency or (ii) assessing a risk of aspiration comprising at least one operation selected from the group consisting of assessing swallowing safety and assessing swallowing efficiency.

6. The method according to claim 1 wherein the steps i)-iii) are performed at least twice, each with one food stuff item selected from the group consisting of water, a foodstuff item with a texture having a nectar-type viscosity, a foodstuff item having a honey-type viscosity, and a foodstuff item having a pudding-type viscosity; and the method comprises associating the performing of each step i) with the one specific foodstuff item thereby providing a sequence of foodstuff items that is to be provided to the subject in step a) or using different algorithms for each of water, a foodstuff item with a texture having a nectar-type viscosity, a foodstuff item having a honey-type viscosity, and a foodstuff item having a pudding-type viscosity.

7. The method of claim 6 wherein, in the sequence of foodstuff items to be provided, at least two of the foodstuff items are the same.

8. The method of claim 1 , wherein, within each sequence of foodstuffs to be provided to the subject, none of the foodstuffs are provided to the subject twice.

9. The method of claim 1 , wherein one of the at least three foodstuffs of step i) in step a) is water.

10. The method of 1 , wherein a sequence of the at least three foodstuff items is water, a foodstuff item with a texture having a nectar-type viscosity, a foodstuff item having a honey-type viscosity, and a foodstuff item having a pudding-type viscosity, and wherein the water is first in the sequence of the at least three foodstuff items.

11. A kit comprising:

a processing unit configured for assessing characteristics of a swallowing process in a subject;

a sensor capable of detecting throat vibrations, the sensor adapted to detect vibrations externally or internally on a throat of the subject; and

a thickening agent formulated for thickening a fluid for producing at least one item of foodstuff with a defined texture for intake by the subject,

wherein the processing unit after training is capable of outputting an indication of the characteristics of the swallowing process in the subject, and

wherein the processing unit is capable of accepting from the sensor a sensor signal associated with the intake of an item of tagged foodstuff by the subject, the processing unit configured to:

a) perform at least once the following steps:

i) provide to the subject at least three items of tagged foodstuff for oral intake, each of the at least three items of tagged foodstuff are selected from the group consisting of water, a foodstuff item with a texture having a nectar-type viscosity of 51 mPa s to 350 mPa s, a foodstuff item having a honey-type viscosity of 301 mPa s to 1750 mPa s, and a foodstuff item having a pudding-type viscosity of more than 1750 mPa s,

ii) assess the characteristics of the swallowing process in the subject resulting from the intake of at least two items of tagged foodstuff by the subject with video fluoroscopy; and iii) train the processing unit using the results of step i) and ii), the training comprising determining the sensor signal resulting from the throat vibrations resulting from the intake of at least two items of tagged foodstuff by the subject, and forwarding the sensor signal to the processing unit, the processing unit is operated by a computer executable software, and the computer executable software provides an assessment of the characteristics of the swallowing process, and the assessment from the computer executable software results from correlating the sensor signal with the assessment resulting from the video fluoroscopy,

repeating steps i) to iii) at least once to improve correlation between an assessment of the characteristics of the swallowing process of the subject resulting from the sensor signal and the assessment of the characteristics of the swallowing process resulting from the video fluoroscopy, wherein the subject is selected from the group consisting of the same subject and a different subject; and

b) thereby obtaining a trained processing unit,

and the processing unit is configured to perform at least one step selected from the group consisting of (A) implementing a neural network and (B) determining parameters of the computer executable software running on and instructing the trained processing unit.

12. A method for assessing characteristics of swallowing in a subject, the method comprising assessing swallowing safety and/or-swallowing efficiency, the assessing comprising the steps:

a) providing to a subject at least three items of foodstuff for oral intake,

b) assessing the characteristics of the swallowing in the subject associated with the oral intake of the at least three items of foodstuff with a kit comprising at least one device comprising a processing unit and a sensor capable of detecting throat vibrations, the kit further comprising a thickening agent for thickening a fluid for producing at least one item of foodstuff with a defined texture for intake by the subject, the processing unit is supplied with a sensor signal from the sensor, the sensor adapted to detect the throat vibrations of the subject externally or internally,

wherein the processing unit after training is capable of outputting an indication of the characteristics of the swallowing process in the subject, and

wherein the processing unit is capable of accepting from the sensor the sensor signal, the sensor signal is associated with the oral intake of the item of tagged foodstuff by the subject, the method comprising:

performing at least once the following steps:

i) providing to the subject at least three items of tagged foodstuff for the oral intake, each of the at least three items of tagged foodstuff are selected from the group consisting of water, a foodstuff item with a texture having a nectar-type viscosity of 51 mPa s to 350 mPa s, a foodstuff item having a honey-type viscosity of 301 mPa s to 1750 mPa s, and a foodstuff item having a pudding-type viscosity of more than 1750 mPa s,

ii) assessing the characteristics of the swallowing process in the subject resulting from the oral intake of at least two items of tagged foodstuff by the subject with video fluoroscopy; and

using the results of step i) and ii) for the training of the processing unit, the training comprising determining the sensor signal resulting from the throat vibrations resulting from the intake of at least two items of tagged foodstuff by the subject, and forwarding the sensor signal to the processing unit, the processing unit is operated by a computer executable software, and the computer executable software provides an assessment of the characteristics of the swallowing process, and the assessment from the computer executable software results from correlating the sensor signal with the assessment resulting from the video fluoroscopy,

repeating steps i) to iii) at least once to improve correlation between an assessment of the characteristics of the swallowing process of the subject resulting from the sensor signal and the assessment of the characteristics of the swallowing process resulting from the video fluoroscopy, wherein the subject is selected from the group consisting of the same subject and a different subject; and

thereby obtaining a trained processing unit,

wherein the method comprises at least one step selected from the group consisting of (A) implementing a neural network and (B) determining parameters of the computer executable software running on and instructing the trained processing unit.

13. The method of claim 12 , wherein at least one of the at least three different defined items of tagged foodstuffs is water.

14. The method of claim 12 wherein the subject is provided with an additional item of tagged foodstuff that has a different texture than the at least three items of foodstuff previously used.

15. The method of claim 12 , wherein a first item of the at least three items of tagged foodstuff provided in step a) is water.