Automated foam detection

A computer-implemented method for controlling a bioprocess in a vessel, the method comprising:

• receiving a sample image of at least a portion of the vessel comprising the liquid-gas interface;
• classifying the sample image between a first class and at least one second class, the first class and at least one second class being associated with different amounts of foam on the surface of the liquid, wherein the classifying is performed by a deep neural network classifier that has been trained using a plurality of training images of at least a portion of a vessel comprising a liquid-gas interface, wherein the plurality of training images comprise at least some images that differ from each other by one or more of:
• the location of the liquid-gas interface on the image, the polar and/or azimuthal angle at which the liquid-gas interface is viewed on the image, and the light intensity or colour temperature of the one or more light sources that illuminated the imaged portion of the vessel when the image was acquired;
• and
• sending a first signal to an effector device if the sample image is classified in the second class or a first one of a plurality of second and further classes, wherein the effector device is selected from an antifoam agent dispensing system, an agitator system, an aeration system, a foam removal system and a foam destruction system;
• and
• repeating the steps of receiving a sample image, classifying the sample image and sending a signal to an effector device if the sample image is classified in the second class or a first one of a plurality of second and further classes, after a predetermined period of time has elapsed since receiving the preceding image.

A system for detecting foam on the surface of a liquid medium contained in a vessel of a bioreactor, the system including:

• at least one processor;
• and
• at least one non-transitory computer readable medium containing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
• receiving a sample image of at least a portion of the vessel comprising the liquid-gas interface;
• classifying the sample image between a first class and at least one second class, the first class and at least one second class being associated with different amounts of foam on the surface of the liquid, wherein the classifying is performed by a deep neural network classifier that has been trained using a plurality of training images of at least a portion of a vessel comprising a liquid-gas interface, wherein the plurality of training images comprise at least some images that differ from each other by one or more of:
• the location of the liquid-gas interface on the image, the polar and/or azimuthal angle at which the liquid-gas interface is viewed on the image, and the light intensity or colour temperature of the one or more light sources that illuminated the imaged portion of the vessel when the image was acquired;
• and
• responsive to classifying the sample image in the at least one second class, sending a signal to an effector device or a warning message to a user interface, wherein the effector is an antifoam agent dispensing system, an agitator system, an aeration system, a foam removal system, or a foam destruction system.

One or more non-transitory computer readable media comprising instructions that, when executed by at least one processor, cause the at least one processor to perform a method for detecting foam on the surface of a liquid medium contained in a vessel of a bioreactor, the method including the steps of:

• receiving a sample image of at least a portion of the vessel comprising the liquid-gas interface;
• classifying the sample image between a first class and at least one second class, the first class and at least one second class being associated with different amounts of foam on the surface of the liquid, wherein the classifying is performed by a deep neural network classifier that has been trained using a plurality of training images of at least a portion of a vessel comprising a liquid-gas interface, wherein the plurality of training images comprise at least some images that differ from each other by one or more of:
• the location of the liquid-gas interface on the image, the polar and/or azimuthal angle at which the liquid-gas interface is viewed on the image, and the light intensity or colour temperature of the one or more light sources that illuminated the imaged portion of the vessel when the image was acquired;
• and
• responsive to classifying the sample image in the at least one second class, sending a signal to an effector device or a warning message to a user interface, wherein the effector is an antifoam agent dispensing system, an agitator system, an aeration system, a foam removal system, or a foam destruction system.