Systems and methods for contact lens fitting

Abstract

A method for fitting a contact lens may include: receiving, from a first user device and at a computing device associated with a contact provider, a response to a questionnaire; receiving, from a second user device, clinical data associated with a contact lens wearer; providing input to a AI trained model (Machine Learning Model), where the input may include the responses to the questionnaire and the clinical data, and where the trained model is trained to receive an input response to the questionnaire and input clinical data and infer a contact lens recommendation; receiving, from the trained model, a contact lens recommendation associated with the contact lens wearer based on the responses to the questionnaire and the clinical data; and transmitting, to the first user device, an indication of the contact lens recommendation.

Description

SYSTEMS AND METHODS FOR CONTACT LENS FITTINGBACKGROUND

[0001] Assessing fit of a multifocal contact lens prior to use of the contact lens can be difficult. Various factors may influence how appropriate a particular contact lens is for a particular contact lens wearer. According to one study, multifocal contact lens wearers were satisfied with their contact lens fit about 65% on a first trial, 93% on a second trial, and 100% on a third trial.

[0002] Improvements are needed.SUMMARY

[0003] A system of one or more computers can be configured to perform particular operations or actions by virtue of having software, firmware, hardware, or a combination of them installed on the system that in operation causes or cause the system to perform the actions. One or more computer programs can be configured to perform particular operations or actions by virtue of including instructions that, when executed by data processing apparatus, cause the apparatus to perform the actions.

[0004] In one general aspect, a method may include receiving, from a first user device and at a computing device associated with a contact provider, a response to a questionnaire, where the response to the questionnaire may include information about a lifestyle associated with a contact lens wearer, where the response to the questionnaire is associated with a contact lens purchase, and where the response to the questionnaire is received prior to the contact lens wearer arriving at a premises associated with the contact lens provider. The method may also include receiving, from a second user device, clinical data (e.g., information, etc.) associated with the contact lens wearer. The method may furthermore include providing input to a trained model, where the input may include the responses to the questionnaire and the clinical data, and where the trained model is trained to receive an input response to the questionnaire and input clinical data and infer a contact lens recommendation. The method may in addition include receiving, from the trained model, a contact lens recommendation associated with the contact lens wearer based on the responses to the questionnaire and the clinical data. The method may moreover include transmitting, to the first user device, an indication of the contact lens recommendation. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0005] Implementations may include one or more of the following features. The method where the clinical data may include refractive data. The method where the clinical data may include sensitivity data. The method where the indication of the contact lens recommendation may include a quick-response (QR) code. The method where the indication of the contact lens recommendation may include a Short Message Service (SMS) message. The method where after the contact lens purchase is completed, transmitting a consumer survey to the first user device. The method may include: receiving, from the first user device, a response to the consumer survey; and updating the Al trained model (Machine Learning Model) based on the response to the questionnaire, the clinical data, the contact lens recommendation, and the response to the consumer survey. The method may include: in response to receiving the response to the questionnaire from the first user device, transmitting, to the second user device, a request for the clinical data associated with the contact lens wearer. The method where the clinical data is automatically transmitted in response to the request for the clinical data. The method where at least a portion of the responses to the questionnaire are automatically created by the first user device. The method where at least a portion of the responses to the questionnaire are automatically created by the first user device based on communication with one or more sensors associated with the first user device. The method where at least a portion of the responses to the questionnaire are automatically created by the first user device based on communication with one or more wearable devices in communication with the first user device. The method where at least a portion of the responses to the questionnaire are automatically created by the first user device based on at least one previous input received for a form. Implementations of the described techniques may include hardware, a method or process, or a computer tangible medium.BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Fig. 1 shows an example environment in which the systems and methods described herein may operate.

[0007] Fig. 2 shows a representation of operational flow of an example method described herein.

[0008] Fig. 3 shows an example graphical user interface (GUI) screen showing a questionnaire associated with the systems and methods described herein.

[0009] Fig. 4 shows a flowchart for an example method described herein.

[0010] Fig. 5 shows a block diagram representing an example implementation of the system and methods described herein.DETAILED DESCRIPTION

[0011] The present disclosure relates to systems and methods for fitting a contact lens. The present disclosure allows a contact lens wearer to transmit lifestyle information via an application and receive a contact lens recommendation via the application. The lifestyle information may be automatically captured, such as through a smartphone and / or a wearable computing device capturing activity, biometric, and / or environmental information. Environmental information may include location, indoor or outdoor, lighting in the environment, etc. This type of lifestyle information capture is more accurate, more granular, and more dynamic than a typical questionnaire a contact lens wearer will typically fill out at a physician’s waiting room.

[0012] The lifestyle information may be transmitted from a user device associated with the contact lens wearer via a network to a server. The lifestyle information may be transmitted via an Application Programming Interface (API) call. The server may cause clinical data to be retrieved from a computing device associated with a physician. A request for the clinical data may be sent via an API call and the clinical data may be returned to the server via an API response. The server may cause the lifestyle information and the clinical data to be inputted into a trained model and receive as output a contact lens recommendation. The server may transmit the contact lens recommendation to the user device associated with the contact lens wearer. The contact lens recommendation may be transmitted via an API call and / or response. This type of real-time contact lens recommendation informs the contact lens wearer of contact lens recommendations prior to arriving at a physician’s location, informing the contact lens wearer of if a new prescription is recommended or not prior to a visit to a physician.

[0013] The present disclosure relates to systems and methods for fitting a contact lens. As an example, methods may comprise receiving, from a first user device and at a computing device associated with a contact lens provider, a response to a questionnaire, wherein the response to the questionnaire comprises information about a lifestyle associated with a contact lens wearer, wherein the response to the questionnaire is associated with a contact lens purchase, and wherein the response to the questionnaire is received prior to the contact lens wearer arriving at a premises associated with the contact lens provider. Methods may further comprise receiving, from a second user device, clinical data associated with the contact lens wearer. Input (e.g., comprising at least a portion of the responses to the questionnaire and the clinical data) may be provided to a trained model, wherein the trainedmodel is trained to receive an input response to the questionnaire and input clinical data and infer a contact lens recommendation. A contact lens recommendation associated with the contact lens wearer based on the responses to the questionnaire and the clinical data may be received from the model. An indication of the contact lens recommendation may be transmitted, for example to the first user device. Based on the contact lens recommendation parameters of a contact lens manufacturing process may be automatically adjusted to produce a customized contact lens for the contact lens wearer.

[0014] Methods may comprise receiving, from a first user device and at a computing device associated with a contact lens provider, a first electrical signal indicative of a dataset via a first application programming interface (API) call. The dataset may comprise a response to a questionnaire. The response to the questionnaire may comprise information about a lifestyle associated with the contact lens wearer. The response to the questionnaire may be associated with a contact lens purchase. The response to the questionnaire may be received prior to the contact lens wearer arriving at a premises associated with the contact lens provider. Methods may comprise receiving, from a second user device and at the computing device associated with the contact lens provider, a second electrical signal via a second API call clinical data associated with the contact lens wearer. Methods may comprise providing input to a trained model via a third electrical signal, wherein the input comprises at least a portion of the responses to the questionnaire and the clinical data, and wherein the trained model is trained to receive an input response to the questionnaire and input clinical data and infer a contact lens recommendation. Methods may comprise receiving, from the trained model, a fourth electrical signal indicative of a contact lens recommendation associated with the contact lens wearer based on the responses to the questionnaire and the clinical data. Methods may comprise transmitting, to the first user device, a fifth electrical signal indicative of an indication of the contact lens recommendation via a third API call. Based on the contact lens recommendation parameters of a contact lens manufacturing process may be automatically adjusted to produce a customized contact lens for the contact lens wearer.

[0015] A computer-implemented method for improving contact lens fitting accuracy and manufacturing efficiency may comprise receiving, by a processor of a computing device associated with a contact lens provider, a digital response to an electronic questionnaire from a first user device, wherein the digital response comprises information about a lifestyle associated with a contact lens wearer. Methods may comprise receiving, by the processor, digital clinical data associated with the contact lens wearer from a second user device. Methods may comprise preprocessing, by the processor, the digital response and the digitalclinical data to generate prep-processed (e.g., normalized) input data, inputting, by the processor, the pre-processed input data into a trained machine learning model, wherein the trained machine learning model has been trained on a dataset of historical contact lens fittings and outcomes. Methods may comprise generating, by the trained machine learning model, a customized contact lens recommendation based on the normalized input data, wherein the customized contact lens recommendation includes specific parameters for manufacturing a contact lens. Methods may comprise transmitting, by the processor, the customized contact lens recommendation to a contact lens manufacturing system. Methods may comprise automatically configuring, by the contact lens manufacturing system, one or more manufacturing devices based on the specific parameters in the customized contact lens recommendation. Methods may comprise manufacturing, by the configured manufacturing devices, a customized contact lens according to the specific parameters.

[0016] Fig. 1 shows an example environment in which the systems and methods described herein may operate. The environment may include a consumer user device 110, a contact provider server 120, an eye care provider user device 130, a trained model, and a network 150.

[0017] The consumer user device 110 may comprise a laptop, smart phone, desktop, or any computing device configured to operate in the environment described herein. The consumer user device 110 may comprise an application. The application may be stored in memory of the consumer user device 110 and executed by one or more processors associated with the consumer user device 110. The application may be an application associated with a contact provider, such as a contact provider operating the contact provider server 120. The application may be a web browser configured to communicate with and access data accessible on the contact provider server 120. The application may use the network 150 to facilitate communication between the consumer user device 110 and the contact provider server 120. The application may comprise an application to facilitate a contact lens purchase for a consumer from the contact provider.

[0018] The contact provider server 120 may comprise one or more computing devices. The contact provider server 120 may comprise a cloud computing environment. The contact provider server 120 may comprise an Application Programming Interface (API) for creating, updating, accessing, and deleting consumer data. Consumer data may comprise a response to a questionnaire. Consumer data may comprise clinical data. Consumer data may comprise contact lens data. Consumer data may comprise consumer survey data.

[0019] The eye care provider user device 130 may comprise a laptop, smart phone, desktop, or any computing device configured to operate in the environment described herein. The eye care provider user device 130 may comprise an application. The application may be stored in memory of the eye care provider user device 130 and executed by one or more processors associated with the eye care provider user device 130. The application may be an application associated with a contact provider, such as a contact provider operating the contact provider server 120. The application may be a web browser configured to communicate with and access data accessible on the contact provider server 120. The application may use the network 150 to facilitate communication between the eye care provider user device 130 and the contact provider server 120. The application may comprise an application to facilitate an eye care provider to provide clinical data on behalf of a consumer to facilitate a contact lens purchase.

[0020] The trained model 140 may be configured to take as input a response to a questionnaire and clinical data. The trained model 140 may be configured to output an inferred contact lens recommendation. Historical input and output sets may be grouped with associated consumer surveys to form a grouping of input / output results. The grouping of input / output results may be used to further train the trained model 140. Although shown as distinct and accessible via the network 150, in some embodiments the trained model 140 may reside within or be in direct communication with one of the other computing devices, such as the contact provider server 120.

[0021] Responses to a questionnaire, clinical data, and / or responses to a consumer survey may be weighted. For example, responses to a first particular question (of a questionnaire and / or a consumer survey) may be a better indicator of a contact recommendation than responses to a second particular question; in such a situation, responses to the first particular question may be weighted more heavily than responses to the second particular question. As another example, a first response to a particular question may be a better indicator of a contact recommendation than a second response to the particular question; in such a situation, feedback with the first response to the particular question may weigh the particular question more heavily than feedback with the second response to the particular question.

[0022] As an example, the trained model 140 may be trained to heavily weight the following fields corresponding to responses to questions in a questionnaire: 1) CurrentDistance (From what distance a consumer spends most the time reading now (“Long distance’7“Intermediate distance - Computer’7“Near distance - cell phone”)), 2) WantDistance (What reading distance does the consumer wish to enhance (“Long distance’7“Intermediate distance -Computer’7“Near distance - cell phone”)), 3) Group_SecondActivity (What activity does the consumer want to do the second most while wearing multifocal lenses (“Daytime driving’7“Nighttime driving’7“Cell phone’7“Computer’7“Other near work’7“Indoor activity’7“Outdoor activity” / “Socia / “Shopping’7“other”)), 4) Group_Addition (Additional power for near vision of a multifocal lens (“Low’7“Med’7“High”)), 5) Group_Age: The consumer’s age (various age ranges or options such as <40, 40-45, 46-50, 51-55, 56-60, OIOS, 66-70])), 6) Group_Sensitvity (The consumer’s sensitivity to changes in diopter, measured during eye test ([1-Not sensitive; 10 - Very sensitive)), 7) Group_FirstActivity (What activity does the consumer want to do the most while wearing multifocal lenses (“Daytime driving” / “Nighttime driving” / “Cell phone ” / “Computer” / “Other near work” / “Indoor activity” / “Outdoor activity” / “Social” / “Shopping” / “other”)), and Group_NearCorrection (what vision correction method does the consumer usually use for near vision (“Single vision - Spectacle” / “Single vision - Contact lens” / “Multifocal - Contant lens” / “Progressive - Spectacles” / “No vision correction” / “other”). Other pre-questionnaire prompts may be used. Other ranges or responses may be provided.

[0023] Use of the systems and methods described herein may reveal more correlations to contact lens recommendation beyond the fields and predefined options for fields initially used. In such a scenario, fields or predefined options for fields may be added, removed, merged, split, given more weight, given less weight, or given no weight. For example, if a large number of consumer profiles have a response of “other” coupled with a similar answer for a particular field, and there appears to be a large correlation between the similar answer for the particular field, then another option incorporating the similar answer may be created for a question associated with the particular field. As another example, if analysis shows that a particular predefined option for a field has little to no correlation to a contact lens recommendation, then the particular predefined option may be removed. As another example, if analysis indicates that two adjacent age range options have similar inferences for a contact lens recommendation, then the two adjacent age ranges may be merged into one age range option. As another example, if analysis reveals a bifurcation of contact lens recommendations between two groups within one age range option, then the two groups may be split into two age range options. As another example, if analysis reveals a value in a field comprises a stronger correlation with contact lens recommendation than previously considered, then the weight associated with the field may be increased. As another example, if analysis reveals a value is a field comprises a weaker correlation with contact lens recommendation than previously considered, then the weight associated with the field may be decreased.

[0024] Input into the trained model 140, such as responses to a questionnaire and / or clinical data, may be pre-processed prior to being input into the trained model 140. The preprocessing of input data may comprise pre-processing the input data to create category groups. A category within a category group may be created when analysis of results reveals a meaningful new category within the category group. As an example of a category group, spherical data received from an eye care provider may be grouped into one of five category groups. The five category groups may include “>1.0”, “-0.75 - 1.0”, “-1.5 - -0.75”, “-2.0 - - 1.5”, “< -2.0”. The category groups may be compared with each other to determine a relevance in predicting a contact lens recommendation, and therefore a weight. The preprocessing of input data may comprise normalization logic for numeric data. For example, the questionnaire in Fig. 3 below shows 8 age ranges; pre-processing may comprise regrouping the 8 age ranges into 3 age groups, which may be analyzed for further adjustment based on accuracy results. The pre-processing of input data may comprise normalization logic for textual re-grouping following business analysis.

[0025] The network 150 may facilitate communication between the consumer user device 110, the contact provider server 120, the eye care provider user device 130, and the trained model 140. The network 150 may comprise a local area network, a wide area network, the Internet, etc. At least a portion of the network 150 may comprise a public network. At least a portion of the network 150 may comprise a private network.

[0026] Fig. 2 shows a representation of operational flow of an example method for multifocal fitting described herein. A questionnaire (or pre-questionnaire) may be transmitted to a user device associated with a consumer prior to the consumer visiting a store (e.g., contact lens provider, etc.) in association with a contact lens purchase. The consumer may fill in the questionnaire prior to visiting the store in association with the contact lens purchase. The questionnaire may be received from a computing device associated with the store. The questionnaire may be filled in automatically. The questionnaire may be received via an application. The application may be associated with the store. For example, fields that the consumer has completed in a previous form may be automatically completed. As another example, biometric data requested by the questionnaire may be provided by one or more wearable devices, such as a watch for example, and / or internet of things (loT) devices, such as a scale for example, in communication with the user device. As another example, questions relating to physical activity may be provided by sensors on the user device that detect movement of the phone, for example.

[0027] The computing device associated with the store may receive a response to the questionnaire from the user device. The computing device associated with the store may cause an alert to be delivered to a service provider user device. The service provider may be an eye care provider. The service provider may be an employee of the store. The alert may be transmitted to a particular device associated with the service provider. The alert may be set in an account of a platform. The account may be associated with the service provider. The platform may be associated with the store. The alert may prompt the service provider to enter clinical data associated with the consumer. The clinical data may comprise sensitivity data associated with the consumer. The clinical data may comprise refractive data associated with the consumer. The computing device associated with the store may receive the clinical data associated with the consumer.

[0028] The computing device associated with the store may provide the response to the questionnaire and the clinical data to a trained model. The trained model may provide a contact lens recommendation. The contact lens recommendation may comprise a corrective power. The contact lens recommendation may comprise a contact lens type. The computing device associated with the store may receive the contact lens recommendation from the trained model. The computing device may transmit the contact lens recommendation to the user device associated with the consumer. Transmitting the contact lens recommendation to the user device associated with the consumer may comprise transmitting a quick-response (QR) code. Transmitting the contact lens recommendation to the user device associated with the consumer may comprise transmitting a Short Message Service (SMS) message.

[0029] Fig. 3 shows an example graphical user interface (GUI) screen for a questionnaire (or pre-questionnaire) associated with the systems and methods described herein. The GUI screen may be presented on a user device associated with a consumer prior to the consumer visiting a store (e.g., contact lens provider, etc.) in association with a contact lens purchase. The questionnaire may comprise questions about age, viewing habits, vision needs, and corrective measures taken. A question in the questionnaire may ask a consumer to select from an option of age ranges. A question in the questionnaire may ask a consumer to classify a distance from which the consumer reads most (from the options of, for example, long distances (road signs, etc.), intermediate distance (computer, etc.), and near distance (cell phone, etc.)). A question in the questionnaire may ask a consumer to classify a distance from which the consumer wishes to enhance reading (from the options of, for example, long distances (road signs, etc.), intermediate distance (computer, etc.), near distance (cell phone, etc.), and no specific vision (overall enhancement)). A question in the questionnaire may aska consumer what are the activities the consumer wants to do most, second most, etc. after wearing multifocal lenses. A question in the questionnaire may ask a consumer to classify a darkness associated with activities the consumer wants to do when wearing multifocal lenses, which may include the options of bright (e.g., outdoors during the day, a department store, etc.), normal (e.g., indoor, office, home, etc.), a bit dark (e.g., a dimly lit cafe, etc.), and dark (e.g., outdoors at night, theater, etc.). A question in the questionnaire may ask a consumer for a vision correction method used to see far away. A question in the questionnaire may ask a consumer for a vision correction method used to see near.

[0030] As an illustrative example, an eye care provider may receive the GUI screen in response to a consumer completing a questionnaire associated with an upcoming contact lens purchase. Through the GUI screen, the eye care provider may enter clinical data associated with the consumer. Clinical data may comprise a sensitivity to changes in diopter. For example, the eye care provider may assess how likely a consumer is to notice a change in diopter. The sensitivity to changes in diopter may be measured as a binary (“sensitive” or “not sensitive”). The sensitivity to changes in diopter may be measured as a range (for example, “consumer detected a difference in . 1 diopter”, “consumer detected a difference in .25 diopter but not . 1 diopter”, “consumer detected a difference in .5 diopter but not lower”, etc.). The sensitivity to changes in diopter may be measured on a scale of 1 to 10. Clinical data may comprise refraction data. Refraction data may comprise spherical data, cylindrical data, axis data, and addition power data. Refraction data may comprise visual acuity data. Visual acuity data may comprise monocular data and binocular data. Refraction data may comprise sensory dominant eye data. Once the clinical data has been entered, the response from the questionnaire and the clinical data may be provided to a trained model. The trained model may infer a contact lens recommendation.

[0031] As an example, a GUI screen may be presented to a user for receiving a contact lens recommendation. The contact lens recommendation may provide a summary of lifestyle and visual needs. The summary of lifestyle and visual needs may comprise a type of multifocal correction, such as balanced, near-focused, far-focused, etc. The contact lens recommendation may provide a product type recommendation. The product type recommendation may be tailored to the consumer’s needs, based on factors such as hours worn, hours spent looking at a screen, sensitivity to light (including photophobia), etc. The contact lens recommendation may provide a profile of multifocal contact lens. The profile of multifocal contact lens may comprise a power and type (Uow, Med, or High) of multifocal contact lens inferred to have a highest probability of success for satisfying the consumer. TheGUI screen may provide an option for sharing the contact lens recommendation with the consumer. The contact lens recommendation may be shared with a QR code. The contact lends recommendation may be shared with an SMS message.

[0032] At least one of the example GUI screens may show how a multifocal contact lens works, such as general instructions and general information about multifocal contact lenses. At least one of the example GUI screens may show the summary of lifestyle and visual needs shown to the eye care provider. At least one of the example GUI screens may show the product type recommendation shown to the eye care provider. At least one of the example GUI screens may show how to use a multifocal contact lens, including general user guidance on how to use (insert, remove, etc.) a multifocal lens.

[0033] The results of the consumer survey may aggregate individual results from the consumer survey and summarize aggregation. The individual results from the consumer survey may be grouped with an input (response from the questionnaire and clinical data) and output (contact lens recommendation) of an associated contact lens purchase and the grouping may be used to further train the model. For example, if the consumer survey indicates a positive result, then a tendency of the trained model to infer the associated output based on the associated input is strengthened. However, if the consumer survey indicates a negative result, then a tendency of the trained model to infer the associated output based on the associated input is weakened.

[0034] As described herein, various training data may be acquired. Training data may comprise an input dataset. The input dataset may comprise clinical data and questionnaire data. Clinical data may comprise at least one of consumer fitting data, refraction data, and prescription data. Each input dataset in the training data may comprise a corresponding output data. The corresponding output data may comprise final prescription data for the input dataset.

[0035] The training data may be used to train a model to create a trained model. The trained model may infer output (final prescription data) based on an input dataset (clinical data and questionnaire data). As explained above, the trained model may be updated based on consumer feedback through consumer surveys.

[0036] Feature engineering may be utilized to determine if the trained model should incorporate additional data in the input dataset. Feature engineering may be utilized to determine if the trained model should eliminate data from the input dataset. Feature engineering may be utilized to determine if the trained model should emphasize (weighted more heavily) and / or discount (weighted less heavily) data in the input dataset. Featureengineering may comprise determining which responses and / or questions from the questionnaire and / or the consumer survey should be emphasized and / or discounted. Feature engineering may comprise determining a question from the questionnaire and / or the consumer survey should be eliminated. Feature engineering may comprise determining a new question should be added to the questionnaire and / or the consumer survey. Feature engineering may comprise determining what clinical data should be used by the trained model. Feature engineering may comprise determining what weight features associated with clinical data should be given in the trained model.

[0037] Multifocal (MF) features may include clinical data. The clinical data may include refractive data. The refractive data may include addition (group_addition: 7.1% importance weighting), sphere (group_sphere: 8.6% importance weighting), dominant (is_dominant: .5% importance weighting), sensitivity (group_sensitivity: 8.1% importance weighting), etc. The MF features may include questionnaire and / or the consumer survey data. The questionnaire and / or the consumer survey data may comprise lifestyle data. The lifestyle data may comprise brightness (6.6% importance weighting), age (group_ages: 6.8% importance weighting), far correction (group_farcorrection: 6.3% importance weighting), near correction (group_nearcorrection: 11% importance weighting), first activity (group_firstactivty: 13.7% importance weighting), second activity (group_secondactivity: 16.2% importance weighting), current distance (curdistance: 7.6% importance weighting), wanted distance (wantdistance: 7.6% importance weighting), etc.

[0038] Lifestyle data may comprise data derived from a pre-questionnaire, such as the questionnaire described in Fig. 3. As described above, the trained model may cause weights to be associated with each MF feature. Various weights may be used, as is shown by illustration above. The weights associated with the MF features may be adjusted as the trained model receives feedback from use. As a further example, refractive data (addition, sphere, dominant, sensitivity) may be accorded 20-30% (e.g., 24.3%) of the importance weighting, while lifestyle data (brightness, ages, farcorrection, nearcorrection, firstactivity, secondactivity, curdistance, wantdistance) may be accorded a majority importance weighting of 70-80% (e.g., 75.7%). Other factors and weights may be used.

[0039] Fig. 4 shows a flowchart of an example process 1000. In some implementations, one or more process blocks of Fig. 10 may be performed by a computing device, such as the contact provider server 120 in Fig. 1.

[0040] As shown in Fig. 4, process 1000 may include receiving a response to a questionnaire (block 1002). For example, the contact provider server 120 in Fig. 1 may receive, from theconsumer user device 110 in Fig. 1, a response to a questionnaire. The response to the questionnaire may include information about a lifestyle associated with a contact lens wearer. The response to the questionnaire may be associated with a contact lens purchase. The response to the questionnaire may be received prior to the contact lens wearer arriving at a premises associated with the contact lens provider. At least a portion of the responses to the questionnaire may be automatically created by the first user device. At least a portion of the responses to the questionnaire may be automatically created by the first user device based on communication with one or more sensors associated with the first user device. At least a portion of the responses to the questionnaire may be automatically created by the first user device based on communication with one or more wearable devices in communication with the first user device. At least a portion of the responses to the questionnaire may be automatically created by the first user device based on at least one previous input received for a form.

[0041] As also shown in Fig. 4, process 1000 may include receiving clinical data (block 1004). For example, the contact provider server 120 may receive, from the eye care provider user device 130 in Fig. 1, clinical data associated with the contact lens wearer. The clinical data may comprise refractive data. The clinical data may comprise sensitivity data. In response to receiving the response to the questionnaire in block 1002, a request for the clinical data associated with the contact lens wearer may be transmitted. For example, in response to receiving the response to the questionnaire, the contact provider server 120 may transmit a request for the clinical data associated with the contact lens wearer to the eye care provider user device 130. The clinical data may be automatically transmitted in response to the request for the clinical data.

[0042] As further shown in Fig. 4, process 1000 may include providing input to a trained model (block 1006). For example, the contact provider server 120 may provide input to the trained model 140 in Fig. 1. The input may include the responses to the questionnaire and the clinical data. The trained model may be trained to receive an input response to the questionnaire and input clinical data and infer a contact lens recommendation.

[0043] As also shown in Fig. 4, process 1000 may include receiving a contact lens recommendation (block 1008). For example, the contact provider server 120 may receive, from the trained model 140, a contact lens recommendation associated with the contact lens wearer based on the responses to the questionnaire and the clinical data.

[0044] As further shown in Fig. 4, process 1000 may include transmitting an indication of the contact lens recommendation (block 1100). For example, the contact provider server 120 maytransmit, to the consumer user device 110, an indication of the contact lens recommendation, as described above. The indication of the contact lens recommendation may comprise a quick-response (QR) code. The indication of the contact lens recommendation may comprise a Short Message Service (SMS) message.

[0045] The process 1000 may include transmitting a consumer survey. For example, the contact provider server 120 may transmit a consumer survey to the consumer user device 110 after the contact lens purchase is completed. The process 1000 may include receiving a response to the consumer survey. For example, the contact provider server 120 may receive, from the consumer user device 110, a response to the consumer survey. The process 1000 may include updating the trained model. For example, the contact provider 120 may cause the trained model 140 to be updated based on the questionnaire, the clinical data, the contact lens recommendation, and the response to the consumer survey.

[0046] Although Fig. 4 shows example blocks of process 1000, in some implementations, process 1000 may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in Fig. 4. Additionally, or alternatively, two or more of the blocks of process 1000 may be performed in parallel.

[0047] Fig. 5 shows a block diagram 1100 representing an example implementation of the system and methods described herein. As shown in the block diagram 1100, a consumer 1102 may make a reservation 1104 for a refractive test 1110 with an eye care provider (ECP). The reservation 1104 may be made in person, via a voice communication, via an electronic written communication, etc. In response to the reservation 1104, the patent 1102 may receive a pre-questionnaire 1106 via electronic communication 1108. The electronic communication 1108 may comprise an electronic-mail (e-mail) message, a text message, a link to a website, a message in an application, etc. The pre-questionnaire may be similar to the questionnaire described in Fig. 3. During the refractive test 1110, the ECP may obtain clinical data, including refractive data, associated with the patient 1102.

[0048] A computing device associated with the ECP 1112 may receive the responses to the pre-questionnaire 1106 and the clinical data obtained from the refractive test 1110. The computing device associated with the ECP 1112 may provide the responses to the prequestionnaire 1106 and the clinical data obtained from the refractive test 1110 to an inference server 1114 via an application programming interface (API). The interface server 1114 may make a query 1116 of if a version of a model used by the interface server 1114 is a current version or if the model has been updated. If the model has been updated, then the inference server 1114 may download the current version of the model from a source 1118, such as acloud server, and then return to the query 1116. If the version of the model used by the interface server 1114 is the current version, then the model may make an inference 1120 using the responses to the pre-questionnaire 1106 and the clinical data obtained from the refractive test 1110. The inference 1120 may be translated into one or more recommendations 1122. The one or more recommendations 1122 may include accuracy recommendations for near, balance, far, etc. The inference server 1114 may return the one or more recommendations to the computing device associated with the ECP 1112 via the API.

[0049] Example Clause 1: A method for fitting a contact lens may include: receiving, from a first user device and at a computing device associated with a contact provider, a response to a questionnaire, where the response to the questionnaire may include information about a lifestyle associated with a contact lens wearer, where the response to the questionnaire is associated with a contact lens purchase, and where the response to the questionnaire is received prior to the contact lens wearer arriving at a premises associated with the contact lens provider; receiving, from a second user device, clinical data associated with the contact lens wearer; providing input to a trained model, where the input may include the responses to the questionnaire and the clinical data, and where the trained model is trained to receive an input response to the questionnaire and input clinical data and infer a contact lens recommendation; receiving, from the trained model, a contact lens recommendation associated with the contact lens wearer based on the responses to the questionnaire and the clinical data; and transmitting, to the first user device, an indication of the contact lens recommendation .

[0050] Example Clause 2: The method of Example Clause 1, where the clinical data may include refractive data.

[0051] Example Clause 3: The method of Example Clause 1 or Example Clause 2, where the indication of the contact lens recommendation may include a quick-response (QR) code.

[0052] Example Clause 4: The method of any one of Example Clauses 1-3, where the indication of the contact lens recommendation may include a Short Message Service (SMS) message.

[0053] Example Clause 5: The method of any one of Example Clauses 1-4, where after the contact lens purchase is completed, transmitting a consumer survey to the first user device.

[0054] Example Clause 6: The method of any one of Example Clauses 1-5, further may include: receiving, from the first user device, a response to the consumer survey; and updating the trained model based on the response to the questionnaire, the clinical data, the contact lens recommendation, and the response to the consumer survey.

[0055] Example Clause 7: The method of any one of Example Clauses 1-6, further may include: in response to receiving the response to the questionnaire from the first user device, transmitting, to the second user device, a request for the clinical data associated with the contact lens wearer.

[0056] Example Clause 8: The method of any one of Example Clauses 1-7, where the clinical data is automatically transmitted in response to the request for the clinical data.

[0057] Example Clause 9: The method of any one of Example Clauses 1-8, where at least a portion of the responses to the questionnaire are automatically created by the first user device.

[0058] Example Clause 10: The method of any one of Example Clauses 1-9, where at least a portion of the responses to the questionnaire are automatically created by the first user device based on communication with one or more sensors associated with the first user device.

[0059] Example Clause 11: The method of any one of Example Clauses 1-10, where at least a portion of the responses to the questionnaire are automatically created by the first user device based on communication with one or more wearable devices in communication with the first user device.

[0060] Example Clause 12: The method of any one of Example Clauses 1-11, wherein at least a portion of the responses to the questionnaire are automatically created by the first user device based on at least one previous input received for a form.

[0061] Example Clause 13: The method of any one of Example Clauses 1-12, wherein the clinical data may include sensitivity data.

[0062] What has been described and illustrated herein is an example along with some of its variations. The terms, descriptions and figures used herein are set forth by way of illustration only and are not meant as limitations. Many variations are possible within the spirit and scope of the subject matter, which is intended to be defined by the following claims — and their equivalents — in which all terms are meant in their broadest reasonable sense unless otherwise indicated.

Claims

CLAIMSWhat is claimed is:

1. A method for fitting a contact lens comprising: receiving, from a first user device and at a computing device associated with a contact lens provider, a response to a questionnaire, wherein the response to the questionnaire comprises information about a lifestyle associated with a contact lens wearer, wherein the response to the questionnaire is associated with a contact lens purchase, and wherein the response to the questionnaire is received prior to the contact lens wearer arriving at a premises associated with the contact lens provider; receiving, from a second user device, clinical data associated with the contact lens wearer; providing input to a trained model, wherein the input comprises at least a portion of the responses to the questionnaire and the clinical data, and wherein the trained model is trained to receive an input response to the questionnaire and input clinical data and infer a contact lens recommendation; receiving, from the trained model, a contact lens recommendation associated with the contact lens wearer based on the responses to the questionnaire and the clinical data; and transmitting, to the first user device, an indication of the contact lens recommendation.

2. The method of claim 1, wherein the clinical data comprises refractive data.

3. The method of claim 1, wherein the indication of the contact lens recommendation comprises a quick-response (QR) code.

4. The method of claim 1, wherein the indication of the contact lens recommendation comprises a Short Message Service (SMS) message.

5. The method of claim 1, wherein after the contact lens purchase is completed, transmitting a consumer survey to the first user device.

6. The method of claim 5, further comprising: receiving, from the first user device, a response to the consumer survey; and updating the trained model based on the response to the questionnaire, the clinical data, the contact lens recommendation, and the response to the consumer survey.

7. The method of claim 1, further comprising: in response to receiving the response to the questionnaire from the first user device, transmitting, to the second user device, a request for the clinical data associated with the contact lens wearer.

8. The method of claim 7, wherein the clinical data is automatically transmitted in response to the request for the clinical data.

9. The method of claim 1, wherein at least a portion of the responses to the questionnaire are automatically created by the first user device.

10. The method of claim 1, wherein at least a portion of the responses to the questionnaire are automatically created by the first user device based on communication with one or more sensors associated with the first user device.

11. The method of claim 1, wherein at least a portion of the responses to the questionnaire are automatically created by the first user device based on communication with one or more wearable devices in communication with the first user device.

12. The method of claim 1, wherein at least a portion of the responses to the questionnaire are automatically created by the first user device based on at least one previous input received for a form.

13. The method of claim 1, wherein the clinical data comprises sensitivity data.

14. A method for fitting a contact lens comprising: receiving, from a first user device and at a computing device associated with a contact lens provider, a response to a questionnaire, wherein the response to the questionnaire comprises information about a lifestyle associated with a contact lens wearer, wherein the response to the questionnaire is associated with a contact lens purchase; receiving, from a second user device, clinical data associated with the contact lens wearer, wherein the clinical data comprises one or more of addition, sphere, dominant, or sensitivity; providing input to a trained model, wherein the input comprises at least a portion of the responses to the questionnaire and the clinical data, and wherein the trained model is trained to receive an input response to the questionnaire and input clinical data and infer a contact lens recommendation, wherein at least a portion of the input is weighted;receiving, from the trained model, a contact lens recommendation associated with the contact lens wearer based on the responses to the questionnaire and the clinical data; and transmitting, to the first user device, an indication of the contact lens recommendation.

15. The method of claim 14, wherein the clinical data comprises refractive data.

16. The method of claim 14, wherein the indication of the contact lens recommendation comprises a quick-response (QR) code or Short Message Service (SMS) message.

17. The method of claim 14, wherein after the contact lens purchase is completed, transmitting a consumer survey to the first user device.

18. The method of claim 18, further comprising: receiving, from the first user device, a response to the consumer survey; and updating the trained model based on the response to the questionnaire, the clinical data, the contact lens recommendation, and the response to the consumer survey.

19. The method of claim 14, wherein the response to the questionnaire is weighted more heavily than the clinical data to effect the contact lens recommendation.

20. A method for fitting a contact lens comprising: causing a pre-formed questionnaire to be presented to a user; receiving, from a first user device and at a computing device associated with a contact lens provider, a response to the questionnaire, wherein the response to the questionnaire comprises one or more lifestyle features, wherein the lifestyle features comprise one or more of brightness, first activity, second activity, current distance, or desired distance; receiving, from a second user device, clinical data associated with the user; providing input to a trained model, wherein the input comprises at least a portion of the responses to the questionnaire and the clinical data, and wherein the trained model is trained to receive an input response to the questionnaire and input clinical data and infer a contact lens recommendation, wherein at least a portion of the input is weighted and the weight accorded to the one or more lifestyle features is greater than the weight accorded to the clinical data;receiving, from the trained model, a contact lens recommendation associated with the contact lens wearer based on the responses to the questionnaire and the clinical data; and transmitting, to the first user device, an indication of the contact lens recommendation.

Images