# US 2026/0091383 A1 · Method for producing a roll of membrane units

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0091383 A1
- Document ID: US20260091383A1
- Status: Published application
- Kind: A1
- Application number: 18/993,391
- Filed: 2023-07-14
- Published: 2026-04-02
- Earliest claimed: 2022-07-14
- Applicants: Sartorius Stedim Biotech GmbH
- Inventors: Jonas AUSTERJOST; Alfons KESTING; Simon STUMM; Robert SÖLDNER
- Web page: https://patentgrep.com/patent/US20260091383A1

## CPC

- B01L 3/5023: {with a sample being transported to, and subsequently stored in an absorbent for analysis}
- G05B 19/41875: {characterised by quality surveillance of production}
- G06T 7/001: {using an image reference approach}
- B01L 2200/12: Specific details about manufacturing devices
- B01L 2300/0825: Test strips
- B01L 2300/165: Specific details about hydrophobic, oleophobic surfaces
- G06T 2207/30108: Industrial image inspection
- B01L: CHEMICAL OR PHYSICAL LABORATORY APPARATUS FOR GENERAL USE
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

The invention is directed to a method for producing a roll (1) or sheet of membrane units (2) for a membrane product (3) such as a lateral flow test from a roll (6) or sheet of membrane material (7), by means of a primary production arrangement (9), wherein the roll (6) or sheet of membrane material (7) is processed into a roll (1) or sheet of membrane units (2) in a primary processing routine, wherein in the primary processing routine, for generating the membrane units (2), a fluidic structure (11), in particular a hydrophobic structure, for defining fluid flow through the membrane material (7) is introduced into the membrane material (7) by means of a processing tool (12), wherein an evaluation routine is performed by means of an evaluation arrangement (15) comprising a sensor arrangement (16) and an evaluation control (17). It is proposed that in the evaluation routine, evaluation images (18) of the fluidic structures (11) of the membrane units (2) are generated by means of the sensor arrangement (16) and evaluation data (19) are generated by means of the evaluation control (17) and that the evaluation data (19) represent the deviation in predefined geometrical properties of the fluidic structure (11) in the respective evaluation image (18) with respect to the fluidic structure (11) in a reference image (21).

## Claims

### Claim 1

Method for producing a roll or sheet (1) of membrane units (2) for a membrane product (3) such as a lateral flow test from a roll or sheet of membrane material (7), by means of a primary production arrangement (9),
wherein the roll or sheet of membrane material (7) is processed into a roll or sheet (1) of membrane units (2) in a primary processing routine,
wherein in the primary processing routine, for generating the membrane units (2), a fluidic structure (11) for defining fluid flow through the membrane material (7), in particular a hydrophobic structure, is introduced into the membrane material (7) by means of a processing tool (12), wherein an evaluation routine is performed by means of an evaluation arrangement (15) with a sensor arrangement (16) and an evaluation control (17),
characterized in that in the evaluation routine, evaluation images (18) of the fluidic structures (11) of the membrane units (2) are generated by means of the sensor arrangement (16) and evaluation data (19) are generated by means of the evaluation control (17) and that the evaluation data (19) represent the deviation in predefined geometrical properties of the fluidic structure (11) in the respective evaluation image (18) with respect to the fluidic structure (11) in a reference image (21).

### Claim 2 (depends on 1)

Method according to claim 1, characterized in that the evaluation images (18) each comprise an array of image data, that are each assigned a location within the respective image border (22), preferably, that the image data are pixel data, distance data or the like.

### Claim 3 (depends on 1)

Method according to any claim 1, characterized in that the sensor arrangement (16) comprises at least one sensor (23) to generate the evaluation images (18), preferably, that the at least one sensor (23) is an optical sensor, in particular a camera sensor, a laser sensor or the like.

### Claim 4 (depends on 1)

Method according to claim 1, characterized in that in a reference routine, from the evaluation images (18) of a number of membrane units (2), the reference image (21) is being defined, preferably, that the reference image (21) is being defined based on averaging the evaluation images (18) of the number of membrane units (2).

### Claim 5 (depends on 1)

Method according claim 1, characterized in that the reference image (21) is an artificial image, in particular an image derived from planning data of the primary processing routine, in particular CAD-data, and/or, that the reference image (21) defined from the evaluation images is compared to a reference image derived from planning data of the primary processing routine, in particular CAD-data.

### Claim 6 (depends on 1)

Method according to claim 1, characterized in that the evaluation data (19) represent a quality grade regarding a degree of deviation between the respective evaluation image (18) and the reference image (21), preferably, that the quality grade is defined based on applying a mathematical operation onto the respective evaluation image (18) and the reference image (21), preferably, that the principle of “sum of squared errors” is applied onto the evaluation image (18) and the reference image (21).

### Claim 7 (depends on 1)

Method according to claim 1, characterized in that the evaluation data (19) represent a deviation in the form of a shift and/or a rotation of the fluidic structure (11) in the respective evaluation image (18) with respect to the fluidic structure (11) in the reference image (21).

### Claim 8 (depends on 7)

Method according to claim 7, characterized in that the evaluation data (19) are generated based on aligning the fluidic structure (11) in the respective evaluation image (18) with the fluidic structure (11) in the reference image (21) and thereby determining the shift and/or rotation of the fluidic structure (11) in the respective evaluation image (18) with respect to the fluidic structure (11) in the reference image (21), preferably, that determining the shift and/or rotation of the fluidic structure (11) is performed in an iterative process, preferably based on the greedy principle.

### Claim 9 (depends on 1)

Method according to claim 1, characterized in that the evaluation data (19) represent a deviation in the form of a skewness of the fluidic structure (11) in the respective evaluation image (18) with respect to the fluidic structure (11) in the reference image (21).

### Claim 10 (depends on 1)

Method according to any claim 1, characterized in that the evaluation data (19) represent a deviation in the form of an incompleteness and/or the existence of structural defects of the fluidic structure (11) in the respective evaluation image (18) with respect to the fluidic structure (11) in the reference image (21).

### Claim 11 (depends on 1)

Method according to any claim 1, characterized in that the deviation is marked in the respective evaluation image (18) and/or in the reference image (21) and/or in a separate image, preferably, that the marked image is being displayed via a user interface.

### Claim 12 (depends on 1)

Method according to claim 1, characterized in that in the evaluation routine, raw images (29) of the membrane units (2) are generated by means of the sensor arrangement (16), which represent the evaluation images (18).

### Claim 13 (depends on 12)

Method according to claim 12, characterized in that in the evaluation routine, raw images (29) of the membrane units (2) are generated by means of the sensor arrangement (16), from which the evaluation images (18) are extracted, preferably, that the respective evaluation image (18) is being extracted from the respective raw image (29) by defining a frame (30) within the raw image (29), that is the image border (22) of the evaluation image (18), further preferably, that the image border (22) is a rectangle, which upper and lower boundaries are defined with respect to a first predefined image structure (31) in the raw image (29) and which left and right boundaries are defined with respect to a second predefined image structure (32) in the raw image (29).

### Claim 14 (depends on 1)

Method according to claim 1, characterized in that the evaluation data (29) are being used by the primary production arrangement (9) to adapt the primary processing routine, preferably, that the primary production arrangement (9) is parameterized based on the evaluation data (19), in order to minimize the detected deviation between the respective evaluation image (18) and the reference image (21).

### Claim 15 (depends on 1)

Method for producing a membrane product such as a lateral flow test from a roll or sheet (1) of membrane units (2) by means of a secondary production arrangement (36), which membrane units (2) have been produced according to claim 1, characterized in that the roll or sheet (1) of membrane units (2) is processed in at least one secondary processing routine based on the evaluation data (19).

### Claim 16 (depends on 15)

Method according to claim 15, characterized in that the secondary production arrangement (36) adapts the secondary process routine to the evaluation data (19), in order to support the production of identical membrane products (3).

### Claim 17 (depends on 15)

Method according to claim 15, characterized in that depending on the detected quality grade, representing a maximum allowed degree of deviation, the respective membrane unit (2) may be discarded from further processing.

### Claim 18 (depends on 15)

Method according to claim 15, characterized in that the secondary processing routine includes the secondary process step of cutting the membrane material (7) into single cut-out membrane units (2), each comprising a fluidic structure (11), preferably, that the cutting step is adapted to the detected deviation such that the membrane units (2) are identical in view of the fluidic structure (11) of the respective membrane unit (2).

### Claim 19 (depends on 15)

Method according to claim 15, characterized in that the secondary processing routine includes the secondary process step of charging the membrane material (7) at a predefined location on the membrane unit (2) with bio-active material such as antibodies, preferably, that charging of the membrane material (7) is adapted to the detected deviation, such that the placement of the bio-active material with respect to the fluidic structure (11) is identical for each membrane unit (2).

### Claim 20 (depends on 15)

Method according to claim 15, characterized in that the secondary processing routine includes the secondary process step of inserting the cut-out membrane units (2) into a housing (4), preferably, that inserting the cut-out membrane units (2) is adapted to the detected deviation such that the fluidic structure (11) is aligned to a structure, preferably a window (5) or the like, of the housing (4).

### Claim 21

Method for evaluating a membrane unit (2) for a membrane product (3) such as a lateral flow test, wherein the membrane unit (2) comprises a fluidic structure (11) for defining fluid flow through the membrane material (7), in particular a hydrophobic structure, that has been introduced into the membrane material (7), wherein an evaluation routine is performed by means of an evaluation arrangement (15) with a sensor arrangement (16) and an evaluation control (17),
characterized in that in the evaluation routine, evaluation images (18) of the fluidic structures (11) of the membrane unit (2) are generated by means of the sensor arrangement (16) and evaluation data (19) are generated by means of the evaluation control (17) and that the evaluation data (19) represent the deviation in predefined geometrical properties of the fluidic structure (11) in the evaluation image (18) with respect to the fluidic structure (11) in a reference image (21).

### Claim at position 22

22-23. (canceled)
