Mobile device-based prenatal screening and intervention for perinatal depression

Abstract

An embodiment includes generating, using a trained perinatal depression (PND) screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data of the user, a PND risk score of the user. An embodiment includes determining that the PND risk score of the user is above a threshold score. An embodiment includes generating, using a trained conversational intervention agent executing on the mobile device of the user, responsive to determining that the PND risk score of the user is above the threshold score, an intervention intended to reduce the PND risk score of the user.

Description

Docket No. 124763-0015MOBILE DEVICE-BASED PRENATAL SCREENING AND INTERVENTION FOR PERINATAL DEPRESSIONCROSS-REFERENCE OF RELATED APPLICATION

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 631,042, filed on April 8, 2024, which is incorporated herein in its entirety.TECHNICAL FIELD

[0002] The present disclosure relates generally to a digital health solution for mental health of pregnant women and their partners and in particular to mobile device-based prenatal screening and intervention for perinatal depression.BACKGROUND

[0003] Lack of Behavioral Health Clinicians:

[0004] The shortage of behavioral health clinicians and the presence of extensive waitlists underscore a critical issue in the healthcare system (https: / / www.commonwealthfund.org / publications / explainer / 2023 / may / understanding-us- behavioral-health-workforce-shortage). This issue disproportionately impacts people of color, non-English speakers, and rural patients. Compounding this problem is the significant economic impact of untreated and undiagnosed perinatal depression (PND) on both the mother and child. As used herein, PND refers to both antepartum and perinatal depression in both a mother (i.e., the pregnant person) and a partner or significant other of the pregnant person. On the provider side, there are structural barriers such as compensation that are limiting clinicians’ interest in joining and staying in the field. These barriers are unlikely to change anytime soon and will continue to exacerbate this problem.

[0005] Scripted and Artificial Intelligence (Al) Self-Guided Therapy:

[0006] A chatbot is a software application or user interface hosted on a website designed to have textual or spoken conversations with a user. An Al chatbot typically uses Al-based techniques, such as deep learning, natural language processing or generative artificial intelligence to simulate a human’s behavior as a conversational partner.

[0007] Al chatbots have proven to significantly reduce the symptoms of depression, as measured by the Patient Health Questionnaire (PHQ-9) (https: / / mental.jmir.Org / 2017 / 2 / el9). There is strong evidence that Al is ready to tackle interpersonal, cognitive behavioral, andDocket No. 124763-0015 other types of therapy by providing convenient and accessible support for mental health issues.

[0008] In the US, one in five pregnancies will lead to a maternal mental health disorder and one in ten will lead to a paternal mental health disorder, yet:• 85% of these patients remain undiagnosed and untreated, resulting in human suffering and significant economic costs.• The prevalence of PND has increased by 30% in the last decade alone.• The health-economic consequences of untreated and undiagnosed PND are staggering, estimated at $31,500 per mother-child dyad.• Women of color have more barriers to accessing maternal mental health care, receiving treatment at half the rate of white women.• Maternal mental health is now the leading cause of death for new moms after giving birth.• Paternal perinatal mental health remains largely taboo and neglected.

[0009] Despite being on the front line of PND care, obstetrician-gynecologists (OBGYNs) and pediatricians find themselves unequipped to help due to a lack of mental health training and overwhelming workloads within short appointment windows.

[0010] For those fortunate enough to receive a diagnosis of perinatal depression after the onset of symptoms, a shortage of behavioral health clinicians specializing in PND results in a lengthy waitlist of months or even up to a year in order to receive appropriate mental health care. Even worse, because of gaps in care like this, mental health has become the leading cause of death for new moms.

[0011] The traditional standard of care also relies on screening new moms for PND at the first postpartum appointment after the baby arrives. 40% of patients do not attend this important checkup, which disproportionately impacts women of color and those from low socioeconomic backgrounds.SUMMARY

[0012] Some embodiments of the present disclosure provide a computer-implemented method for mobile device-based prenatal screening and intervention for perinatal depression (PND). The method includes generating, using a trained PND screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data ofDocket No. 124763-0015 the user, a PND risk score of the user; determining that the PND risk score of the user is above a threshold score; and generating, using a trained conversational intervention agent executing on the mobile device of the user, responsive to determining that the PND risk score of the user is above the threshold score, an intervention intended to reduce the PND risk score of the user.

[0013] Some embodiments of the present disclosure provide a non-transitory computer- readable medium storing a program for mobile device-based prenatal screening and intervention for perinatal depression (PND). The program, when executed by a computer, configures the computer to generate, using a trained PND screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data of the user, a PND risk score of the user; determine that the PND risk score of the user is above a threshold score; and generate, using a trained conversational intervention agent executing on the mobile device of the user, responsive to determining that the PND risk score of the user is above the threshold score, an intervention intended to reduce the PND risk score of the user.

[0014] Some embodiments of the present disclosure provide a system for mobile devicebased prenatal screening and intervention for perinatal depression (PND). The system comprises a processor and a non-transitory computer readable medium storing a set of instructions, which when executed by the processor, configure the processor to generate, using a trained PND screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data of the user, a PND risk score of the user; determine that the PND risk score of the user is above a threshold score; and generate, using a trained conversational intervention agent executing on the mobile device of the user, responsive to determining that the PND risk score of the user is above the threshold score, an intervention intended to reduce the PND risk score of the user.BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings, which are included to provide further understanding and are incorporated in and constitute a part of this specification, illustrate disclosed embodiments and together with the description serve to explain the principles of the disclosed embodiments.Docket No. 124763-0015

[0016] FIG. 1 illustrates a network architecture used to implement mobile device-based prenatal screening and intervention for perinatal depression, according to some embodiments.

[0017] FIG. 2 is a block diagram illustrating details of a system for mobile device-based prenatal screening and intervention for perinatal depression, according to some embodiments.

[0018] FIG. 3 depicts a block diagram of an example configuration for mobile devicebased prenatal screening and intervention for perinatal depression, in accordance with an illustrative embodiment.

[0019] FIG. 4 depicts an example of components of a product implementing mobile device-based prenatal screening and intervention for perinatal depression, in accordance with an illustrative embodiment.

[0020] FIG. 5 depicts an example of authentication flows used in mobile device-based prenatal screening and intervention for perinatal depression, in accordance with an illustrative embodiment.

[0021] FIG. 6 depicts an example of a product architecture implementing mobile devicebased prenatal screening and intervention for perinatal depression, in accordance with an illustrative embodiment.

[0022] FIG. 7 depicts a flowchart of an example process for mobile device-based prenatal screening and intervention for perinatal depression, in accordance with an illustrative embodiment.

[0023] In one or more implementations, not all of the depicted components in each figure may be required, and one or more implementations may include additional components not shown in a figure. Variations in the arrangement and type of the components may be made without departing from the scope of the subject disclosure. Additional components, different components, or fewer components may be utilized within the scope of the subject disclosure.DETAILED DESCRIPTION

[0024] In the following detailed description, numerous specific details are set forth to provide a full understanding of the present disclosure. It will be apparent, however, to one ordinarily skilled in the art, that the embodiments of the present disclosure may be practicedDocket No. 124763-0015 without some of these specific details. In other instances, well-known structures and techniques have not been shown in detail so as not to obscure the disclosure.

[0025] One in five pregnant women and one in ten of their partners experience postpartum depression (PPD). Further, although labeled postpartum depression, almost half of cases occur antepartum, i.e., before giving birth. Today, there is no easily accessible way to detect early which pregnant mothers or their partners may develop PND. This prevents early preventive intervention from happening effectively. Moreover, there exists an extremely limited access to care in the US with 30-50% of pregnant mothers not having access to an obstetrical care provider. Together, lack of timely application of known effective interventions such as ROSE protocol (Reach Out, Stay Strong, Essentials for mothers of newborns) and poor access to care hinder prevention and reduction of PND impact on pregnancy outcomes. Thus, there is a need for a proactive, preventive, scalable, and equitably accessible digital health solution for detecting and preventing or reducing PND both during and after pregnancy.

[0026] Embodiments of the present disclosure address the above identified problems by implementing mobile device-based prenatal screening and intervention for perinatal depression, including a machine learning-based method for early detection of an increased risk for PND (AUC=0.91) combined with an artificial intelligence- (large language model) driven scripted gamified chatbot based on ROSE protocol for proactive preventive intervention known to reduce the chance of PND from manifesting by 50% or reduce the severity of PND in the cases when it does occur.

[0027] Embodiments include an engaging and clinically effective method and system of delivering early risk screening and cognitive-behavioral and interpersonal therapies (CBT) virtually and designed to reduce the risk of PND and its severity. Embodiments include an innovative digital health platform that tackles PND even before the symptoms arise. Embodiments implement early engagement, as embodiments are integrated with OBGYN clinics, introducing an application implementing an embodiment during pre-registration to all women. This initial introduction serves as a crucial entry point for pregnant women to learn about the application that raises awareness of PND and its preventive aspects. This in turn helps with clinician burnout because if there is no awareness and necessary resources to address the issue OBGYNs end up spending more time in their scheduled 10 to 15-minute appointments causing burnout issues with prolonged appointments. Embodiments are predictive and personalized, as embodiments identify high-risk women through predictionDocket No. 124763-0015 screening during (e.g., at 22 weeks) pregnancy and offer self-directed therapy, including daily mood tracking and a chatbot for ongoing support. Embodiments offer manageable content, as prevention modules are bite-sized and engaging, for example lasting 5 minutes each for 28 days and providing manageable and consistent engagement. Embodiments provide empathetic support, using gentle reminders and personalized outreach from behavioral health specialists to ensure adherence, prioritizing mothers' well-being. Embodiments empower women to proactively manage their mental health through early awareness, personalized support, and accessible tools.

[0028] In particular, an embodiment generates, using a trained PND screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data of the user, a PND risk score of the user; determines that the PND risk score of the user is above a threshold score; and generates, using a trained conversational intervention agent executing on a mobile device of a user, responsive to determining that the PND risk score of the user is above the threshold score, an intervention intended to reduce the PND risk score of the user.

[0029] An embodiment generates, using a trained PND screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data of the user, a PND risk score of the user. In another embodiment, the PND screening model executes on another device, such as a website or an application executing on a server in a cloud environment, or via a voice-based virtual assistant or user interface. Note that a user includes a pregnant person and a spouse, partner, co-parent, or significant other of a pregnant person.

[0030] Some embodiments implement a PND risk prediction technique incorporating answers to multiple questions, including but not limited to user health history and sociodemographics of a user, but also capable of incorporating input from wearables or ambient biosensors. One embodiment generates a PND risk score using a trained machine learning model, for example, a distributed random forest or logistic regression model, both presently available techniques. Some non-limiting examples of data used in the trained machine learning model are history of depression before pregnancy, history of mental health condition before pregnancy, recent medications for a mental health condition at the initial visit, body mass index (BMI), income or income bracket of a predefined range, age, history of anxiety before pregnancy, education, and discussing with a provider about preparedness for pregnancy before becoming pregnant. Embodiments source user data used in the trainedDocket No. 124763-0015 machine learning model from user-provided data and data extracted from the user’s electronic medical records.

[0031] Some embodiments executing on a device with a text input capability use presently available natural language analysis techniques to generate a PND risk score from collected text data and other data of a user. In some embodiments, collected text and other data of a user includes a free writing portion, in which a user is invited to share how the user has experienced pregnancy over the past month (or another time period), focusing on emotions and feelings. In some embodiments, collected text and other data of a user includes an entry mood screener, in which the user is invited to rate their mood on a scale from “very happy” to “very sad” via a user interface including a slider or numerical rating provided via text entry. In some embodiments, the session includes an instruction to read a positive affirmation and a post-affirmation mood screener, in which the user is invited to rate their mood on a scale from “very happy” to “very sad” via a user interface including a slider or numerical rating provided via text entry. In some embodiments, the session includes depression screen questions that assess how the user has been feeling (e.g., in the past week).

[0032] Some embodiments executing on a device with a microphone use presently available speech analysis techniques to generate a PND risk score from collected speech data and other data of a user. In some embodiments, collected speech and other data of a user includes a free speech portion, in which a user is invited to share how the user has experienced pregnancy over the past month (or another time period), focusing on emotions and feelings. In some embodiments, collected speech and other data of a user includes an entry mood screener, in which the user is invited to rate their mood on a scale from “very happy” to “very sad” via a user interface including a slider or numerical rating provided via voice or text entry. In some embodiments, collected speech and other data of a user includes a voice sample portion, in which the user is instructed to read a positive affirmation aloud. In some embodiments, the session includes a post-affirmation mood screener, in which the user is invited to rate their mood on a scale from “very happy” to “very sad” via a user interface including a slider or numerical rating provided via voice or text entry. In some embodiments, collected speech and other data of a user includes a free speech portion, in which the user is invited to reflect on the state of the world. In some embodiments, the session includes depression screening questions that assess how the user has been feeling (e.g., in the past week).Docket No. 124763-0015

[0033] In one embodiment, the PND risk is predicted for PND requiring medication. In another embodiment, the PND risk is predicted for PND requiring a form of psychotherapy. In some embodiments, the PND risk is predicted for a particular time, such as a particular pregnancy trimester or a particular postpartum time.

[0034] Some embodiments generate a PND risk score as part of a user’s prenatal care, for example at a 22 weeks of pregnancy appointment at a medical clinic with a prenatal caregiver such as an OB GYN. Other embodiments generate a PND risk score as part of a user’s postnatal care.

[0035] Some embodiments report the PND risk score to a user, optionally along with additional information about PND or along with an intervention described elsewhere herein. With the user’s consent, some embodiments communicate a PND risk score to a user’s care provider (for example, primary care provider or OB GYN) via one or several channels, such as secure email. The user’s care provider has the option to refer the user for PND prevention or mitigation care, for example from the user’s existing care provider or another care provider (e.g., a psychologist or other mental health specialist). Some embodiments report the PND risk score along with an explanation of how the PND risk score was generated, to a user, the user’s care giver, or to another application (e.g., for model validation, additional research, or additional model training). Some non-limiting examples of an explanation are names, descriptions, and values of types of data that contributed most, or contributed at all, to generation of the user’s PND risk score.

[0036] Some embodiments are accessible via a clinic user authorization flow in which a clinic user (i.e., a caregiver of a user) receives a link to a user results management website, a clinic user accesses the user results management website which triggers a redirection to a single sign-on (SSO) website, a clinic user logs into the SSO website, a clinic user’s registered email address or other contact method receives a confirmation code, a cloudbased directory and identity management service such as Azure Active Directory (AD) triggers an authorization chain, a clinic user logs in to the user results management website with the option to reduce secure sign on for future log-ins, and the clinic user is able to access Create, Read, Update, Insert, and Delete (CRUID) related operations on data, view a dashboard of user results, and upload additional data. (Azure Active Directory and Azure AD are registered trademarks of Microsoft Corporation in the United States and other countries.) Some embodiments are accessible via a user authorization flow in which a user’s contact method (e.g., email or phone) are derived from clinic data or a form the userDocket No. 124763-0015 completes, the user receives a link to a website for login and registration, the user’s registered email address or other registered contact method receives a confirmation code, a cloud-based directory and identity management service such as Azure AD triggers an authorization chain, the user logs in to a website or application implementing screening and intervention functionality with the option to reduce secure sign on for future log-ins, and the user is able to access a website or application implementing screening and intervention functionality.

[0037] An embodiment determines that the PND risk score of the user is above a threshold score. Responsive to determining that the PND risk score of the user is above the threshold score, an embodiment generates, using a trained conversational intervention agent executing on a mobile device of a user, an intervention intended to reduce the user’s PND risk score. In another embodiment, the trained conversational intervention agent executes on another device, such as a website or an application executing on a server in a cloud environment, or via a voice-based virtual assistant or user interface.

[0038] Some embodiments offer a preventive treatment to the user. In some embodiments, treatment is initiated, for example, before or during pregnancy (e.g., during the first pregnancy trimester) to reduce the risk of PND, in some estimates up to 50% or higher, or to reduce PND’s severity and may continue after delivery for up to 12 months postpartum. In some embodiments, treatment is initiated after the pregnancy ends. In some embodiments, the treatment is based on the clinically validated ROSE protocol (Reach Out, Stay Strong, Essentials for mothers of newborns). With user consent, some embodiments offer an intervention, such as CBT or interpersonal, or other psychological support, to the partner of the pregnant person, also referred to as the significant other, or one or more family members engaged in the shared pregnancy experience and care before, during and after delivery.

[0039] In embodiments, the intervention includes using a virtual conversational perinatal depression prevention agent to solicit a response from the user. One embodiment provides an intervention using a scripted chatbot implementation (a presently available technique), for example based on the ROSE clinical protocol. Another embodiment provides an intervention using a large language model (LLM) that is designed, e.g., via fine-tuning, retrieval augmented generation, prompt engineering or other presently available techniques constraining the performance and behavior to a desired, highly scripted, professional yet conversational and engaging manner. Providing the intervention using a virtual conversational perinatal depression prevention agent or LLM enables a user to access theDocket No. 124763-0015 intervention in a selected language, such as English, Spanish, or Cantonese, and allows for rapid implementation of additional languages. One embodiment also adapts a level of language used in the intervention based on the user’s stated education or the language of that education (e.g., obtained from an intake form or the user’s electronic medical records). For example, a user with only an elementary school education or receiving an intervention in a language the user is less proficient in might be offered a lower level of language (e.g., simplified vocabulary and sentence structure) as compared to a user with a graduate degree receiving an intervention in their native language.

[0040] In embodiments, the agent includes effective therapist characteristics, communication styles, the needs of specific cultural groups, and empathy. In embodiments, performance of the agent is personalized which defines the individual user experience in terms of aspects such as duration and frequency of each session as well as other forms of gamification that drive optimal individual adherence to the virtual psychological support experience, during and after pregnancy and if PND does occur, during the PND to help adhering and completing the treatment. Some non-limiting examples of intervention session times are two minutes, three minutes, and five minutes. Some non-limiting examples of the number of intervention sessions are ten days and 28 days. In addition, some users complete interventions in fewer or more sessions than originally scheduled. Other session lengths and numbers of intervention sessions are also possible and contemplated within the scope of the illustrative embodiments.

[0041] In some embodiments, a session includes one or more of a check-in questionnaire (e.g., to be completed daily), topic-based chatbot content, and meditation exercises. In some embodiments, the check-in questionnaire includes questions from PHQ2, PHQ8, or EDPS, presently available patient health questionnaires.

[0042] One non-limiting example of topic-based chatbot content is, “Did your mother ever share with you what it was like the first couple of months of having you as a baby?” with an option of selecting yes or no. If the user selects yes, an embodiment might respond, “That’s great! It’s helpful to learn from her experiences.” If the user selects no, an embodiment might respond, “That’s okay. Many women don’t talk about that time because it can be so personal or challenging. While some women find the early months of motherhood easier, many experience challenges during the first three months after birth. I want to share some honest realities with you — not to worry you, but to help you feel prepared and avoid feelingDocket No. 124763-0015 guilty if things feel tough at times. Knowing what to expect can make the experience less overwhelming and help you navigate these changes with more confidence.”

[0043] Another non-limiting example of topic-based chatbot content is, “Let’s think about what might change when baby arrives. Sometimes we focus so much on what we're giving up that we forget about what we're gaining. What do you think you might miss about your current life?” If the user responds, “Free time with my husband, losing my pre-pregnancy body, not easy to go out with friends without considering babysitting, less freedom with money,” an embodiment might respond, “That’s a thoughtful way of reflecting on the changes ahead. It’s totally normal to miss some of the things you enjoyed before, like the freedom to spend time with your husband, your body, or being able to go out without extra planning.”

[0044] In some embodiments, an intervention includes an opportunity to create a personalized virtual environment, such as a garden world and gardening activities, where the user picks their child’s birth flower, which the user will “nurture” over time from a seedling to a blossoming plant by completing small daily tasks. This approach aligns with the values of care, patience, and gentle growth, matching the self-compassionate progress embodiments are intended to help support.

[0045] Some embodiments generate an intervention using data from one or more sensors. For example, an embodiment might use a presently available voice analysis technique to measure, from a sample of the user’s speech, biomarkers indicating that the user is depressed or stressed, and tailor an intervention accordingly. For example, if the user is particularly stressed, an intervention might include, “You seem particularly stressed today, would you like me to walk you through a breathing exercise that many folks find calming?” As another example, an embodiment might use one or more biometric sensors to measure a user’s sleep time or sleep quality and offer an intervention such as, “I see you only slept four hours last night, would you like some suggestions to help you get more sleep?” As another example, an embodiment might use one or more biometric sensors (e.g., an accelerometer, a heart rate sensor, a step counter, and the like) to measure a user’s level of physical activity and offer an intervention such as, “I see you got in a walk every day this week - great job!”

[0046] In some embodiments, to further personalize and gamify the experience with the goal of best possible adherence to session completion, upon consent from the user and other participants, an individual social network is created for the user leveraging the initialDocket No. 124763-0015 personal information provided by the user. This network can be drawn upon via publicly available social network data sources. The goal is to use the power of social networks to provide emotional and active support for the pregnant person during their journey, and thus in some embodiments an intervention includes use of the generated social network. For example, users might view each other's progress in a visually appealing and non-intrusive manner through a cohesive visual representation, such as growing flowers, helping to cultivate a sense of unity and reduce feelings of isolation. As another example, users might (with consent) interact with one another (e.g., by "watering" each other’s flowers), serving as a simple yet impactful gesture of encouragement, a subtle form of social interaction designed to foster a sense of camaraderie and strengthen the bonds within the community of expectant mothers. As another example, an embodiment might suggest an intervention a user’s social media contact might perform with the user. At any time during any session, the provider or an emergency hotline, e.g., to prevent harm to the user or others, can be contacted, for example, via audio, video or both or via messaging. In one embodiment, the provider or the emergency hotline may be contacted automatically based on the content of the conversation to preemptively reduce the risk of harm to self or others.

[0047] One embodiment includes one or more adherence protocols to facilitate an attrition- free completion of the psychotherapy or pharmacotherapy for those patients who do go on to develop PND requiring such treatment. These protocols leverage the disclosed individualized gamification features to enable an adherence protocol. Some embodiments monitor user engagement with one or more interventions. If a user does not participate in one or more interventions, does not complete one or more interventions, or an embodiment determines from the user’s answers or other user data that user engagement is below a threshold level, embodiments trigger execution of an adherence protocol. Some embodiments, instead of or in addition to an adherence protocol, engage the user via proactive nudges via email, text message, or using another communication method.

[0048] Some embodiments, during or after completion of a PND intervention protocol, generate an updated PND risk score of the user, using the trained PND screening model. If the updated PND risk score is above a threshold, or is trending upwards, some embodiments adjust the PND intervention protocol, and other embodiments (with user consent) alert a caregiver, partner, or other family member to the updated PND risk score or trend.

[0049] An embodiment includes a postpartum preventive wrap-up session. In one embodiment, this session is based on the ROSE protocol. In some embodiments, the partnerDocket No. 124763-0015 and / or the family members are also involved in the sessions prepared specifically for these family members.

[0050] FIG. 1 illustrates a network architecture 100 used to implement mobile devicebased prenatal screening and intervention for perinatal depression, according to some embodiments. The network architecture 100 may include one or more client devices 110 and servers 130, communicatively coupled via a network 150 with each other and to at least one database 152. Database 152 may store data and files associated with the servers 130 and / or the client devices 110. In some embodiments, client devices 110 collect data, video, images, and the like, for upload to the servers 130 to store in the database 152.

[0051] The network 150 may include a wired network (e.g., fiber optics, copper wire, telephone lines, and the like) and / or a wireless network (e.g., a satellite network, a cellular network, a radiofrequency (RF) network, Wi-Fi, Bluetooth, and the like). The network 150 may further include one or more of a local area network (LAN), a wide area network (WAN), the Internet, and the like. Further, the network 150 may include, but is not limited to, any one or more of the following network topologies, including a bus network, a star network, a ring network, a mesh network, and the like.

[0052] Client devices 110 may include, but are not limited to, laptop computers, desktop computers, and mobile devices such as smart phones, tablets, televisions, wearable devices, head-mounted devices, display devices, and the like.

[0053] In some embodiments, the servers 130 may be a cloud server or a group of cloud servers. In other embodiments, some or all of the servers 130 may not be cloud-based servers (i.e., may be implemented outside of a cloud computing environment, including but not limited to an on-premises environment), or may be partially cloud-based. Some or all of the servers 130 may be part of a cloud computing server, including but not limited to rackmounted computing devices and panels. Such panels may include but are not limited to processing boards, switchboards, routers, and other network devices. In some embodiments, the servers 130 may include the client devices 110 as well, such that they are peers.

[0054] FIG. 2 is a block diagram illustrating details of a system 200 for mobile devicebased prenatal screening and intervention for perinatal depression, according to some embodiments. Specifically, the example of FIG. 2 illustrates an exemplary client device 110-1 (of the client devices 110) and an exemplary server 130-1 (of the servers 130) in the network architecture 100 of FIG. 1.Docket No. 124763-0015

[0055] Client device 110-1 and server 130-1 are communicatively coupled over network 150 via respective communications modules 202-1 and 202-2 (hereinafter, collectively referred to as “communications modules 202”). Communications modules 202 are configured to interface with network 150 to send and receive information, such as requests, data, messages, commands, and the like, to other devices on the network 150.Communications modules 202 can be, for example, modems or Ethernet cards, and / or may include radio hardware and software for wireless communications (e.g., via electromagnetic radiation, such as radiofrequency (RF), near field communications (NFC), Wi-Fi, and Bluetooth radio technology).

[0056] The client device 110-1 and server 130-1 also include a processor 205-1, 205-2 and memory 220-1, 220-2, respectively. Processors 205-1 and 205-2, and memories 220-1 and 220-2 will be collectively referred to, hereinafter, as “processors 205,” and “memories 220.” Processors 205 may be configured to execute instructions stored in memories 220, to cause client device 110-1 and / or server 130-1 to perform methods and operations consistent with embodiments of the present disclosure.

[0057] The client device 110-1 and the server 130-1 are each coupled to at least one input device 230-1 and input device 230-2, respectively (hereinafter, collectively referred to as “input devices 230”). The input devices 230 can include a mouse, a controller, a keyboard, a pointer, a stylus, a touchscreen, a microphone, voice recognition software, a joystick, a virtual joystick, a touch-screen display, and the like. In some embodiments, the input devices 230 may include cameras, microphones, sensors, and the like. In some embodiments, the sensors may include touch sensors, acoustic sensors, inertial motion units and the like.

[0058] The client device 110-1 and the server 130-1 are also coupled to at least one output device 232-1 and output device 232-2, respectively (hereinafter, collectively referred to as “output devices 232”). The output devices 232 may include a screen, a display (e.g., a same touchscreen display used as an input device), a speaker, an alarm, and the like. A user may interact with client device 110-1 and / or server 130-1 via the input devices 230 and the output devices 232.

[0059] Memory 220-1 may further include an application 222, configured to execute on client device 110- 1 and couple with input device 230-1 and output device 232-1 , and implement mobile device-based prenatal screening and intervention for perinatal depression. The application 222 may be downloaded by the user from server 130-1, and / or may beDocket No. 124763-0015 hosted by server 130-1. The application 222 may include specific instructions which, when executed by processor 205-1, cause operations to be performed consistent with embodiments of the present disclosure. In some embodiments, the application 222 runs on an operating system (OS) installed in client device 110-1. In some embodiments, application 222 may run within a web browser. In some embodiments, the processor 205-1 is configured to control a graphical user interface (GUI) (e.g., spanning at least a portion of input devices 230 and output devices 232) for the user of client device 110-1 to access the server 130-1.

[0060] In some embodiments, memory 220-2 includes an application engine 232. The application engine 232 may be configured to perform methods and operations consistent with embodiments of the present disclosure. The application engine 232 may share or provide features and resources with the client device 110-1, including data, libraries, and / or applications retrieved with application engine 232 (e.g., application 222). The user may access the application engine 232 through the application 222. The application 222 may be installed in client device 110-1 by the application engine 232 and / or may execute scripts, routines, programs, applications, and the like provided by the application engine 232.

[0061] Memory 220-1 may further include an application 223, configured to execute in client device 110-1. The application 223 may communicate with service 233 in memory 220-2 to provide mobile device-based prenatal screening and intervention for perinatal depression. The application 223 may communicate with service 233 through API layer 240, for example.

[0062] FIG. 3 depicts a block diagram of an example configuration for mobile devicebased prenatal screening and intervention for perinatal depression, in accordance with an illustrative embodiment. Application 222 is the same as application 222 in FIG. 2.

[0063] Application 222 implements mobile device-based prenatal screening and intervention for perinatal depression, including a machine learning-based method for early detection of an increased risk for PND (AUC=0.91) combined with an artificial intelligence- (large language model) driven scripted gamified chatbot based on ROSE protocol for proactive preventive intervention known to reduce the chance of PND from manifesting by 50% or reduce the severity of PND in the cases when it does occur.

[0064] Application 222 includes an engaging and clinically effective method and system of delivering early risk screening and cognitive-behavioral and interpersonal therapies (CBT) virtually and designed to reduce the risk of PND and its severity. Application 222 includes a digital health platform that tackles PND even before the symptoms arise. Application 222Docket No. 124763-0015 implements early engagement, as implementations of application 222 are integrated with OBGYN clinics, introducing application 222 during pre-registration to all women. This initial introduction serves as a crucial entry point for pregnant women to learn about the application that raises awareness of PND and its preventive aspects. This in turn helps with clinician burnout because if there is no awareness and necessary resources to address the issue OBGYNs end up spending more time in their scheduled 10 to 15-minute appointments causing burnout issues with prolonged appointments. Application 222 is predictive and personalized, identifying high-risk women through prediction screening during (e.g., at 22 weeks) pregnancy and offering self-directed therapy, including daily mood tracking and a chatbot for ongoing support. Application 222 offers manageable content, as prevention modules are bite-sized and engaging, for example lasting 5 minutes each for 28 days and providing manageable and consistent engagement. Application 222 provides empathetic support, using gentle reminders and personalized outreach from behavioral health specialists to ensure adherence, prioritizing mothers' well-being. Application 222 empowers women to proactively manage their mental health through early awareness, personalized support, and accessible tools.

[0065] Screening module 310 generates, using a trained perinatal depression (PND) screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data of the user, a PND risk score of the user. In another implementation of module 310, the PND screening model executes on another device, such as a website or an application executing on a server in a cloud environment.

[0066] Some implementations of module 310 implement a PND risk prediction technique incorporating answers to multiple questions, including but not limited to user health history and sociodemographics of a user, but also capable of incorporating input from wearables or ambient biosensors. One implementation of module 310 generates a PND risk score using a trained machine learning model, for example, a distributed random forest or logistic regression model, both presently available techniques. Some non-limiting examples of data used in the trained machine learning model are history of depression before pregnancy, history of mental health condition before pregnancy, recent medications for a mental health condition at the initial visit, body mass index (BMI), income or income bracket of a predefined range, age, history of anxiety before pregnancy, education, and discussing with a provider about preparedness for pregnancy before becoming pregnant. Implementations ofDocket No. 124763-0015 module 310 source user data used in the trained machine learning model from user- provided data and data extracted from the user’s electronic medical records.

[0067] Some implementations of module 310 executing on a device with a text input capability use presently available natural language analysis techniques to generate a PND risk score from collected text data and other data of a user. In some implementations of module 310, collected text and other data of a user includes a free writing portion, in which a user is invited to share how the user has experienced pregnancy over the past month (or another time period), focusing on emotions and feelings. In some implementations of module 310, collected text and other data of a user includes an entry mood screener, in which the user is invited to rate their mood on a scale from “very happy” to “very sad” via a user interface including a slider or numerical rating provided via text entry. In some implementations of module 310, the session includes an instruction to read a positive affirmation and a post-affirmation mood screener, in which the user is invited to rate their mood on a scale from “very happy” to “very sad” via a user interface including a slider or numerical rating provided via text entry. In some implementations of module 310, the session includes depression screening questions that assess how the user has been feeling (e.g., in the past week).

[0068] Some implementations of module 310 executing on a device with a microphone use presently available speech analysis techniques to generate a PND risk score from collected speech data and other data of a user. In some implementations of module 310, collected speech and other data of a user includes a free speech portion, in which a user is invited to share how the user has experienced pregnancy over the past month (or another time period), focusing on emotions and feelings. In some implementations of module 310, collected speech and other data of a user includes an entry mood screener, in which the user is invited to rate their mood on a scale from “very happy” to “very sad” via a user interface including a slider or numerical rating provided via voice or text entry. In some implementations of module 310, collected speech and other data of a user includes a voice sample portion, in which the user is instructed to read a positive affirmation aloud. In some implementations of module 310, the session includes a post-affirmation mood screener, in which the user is invited to rate their mood on a scale from “very happy” to “very sad” via a user interface including a slider or numerical rating provided via voice or text entry. In some implementations of module 310, collected speech and other data of a user includes a free speech portion, in which the user is invited to reflect on the state of the world. In someDocket No. 124763-0015 implementations of module 310, the session includes depression screen questions that assess how the user has been feeling (e.g., in the past week).

[0069] In one implementation of module 310, the PND risk is predicted for PND requiring medication. In another implementation of module 310, the PND risk is predicted for PND requiring a form of psychotherapy. In some implementations of module 310, the PND risk is predicted for a particular time, such as a particular pregnancy trimester or a particular postpartum time.

[0070] Some implementations of module 310 generate a PND risk score as part of a user’s prenatal care, for example at a 22 weeks of pregnancy appointment at a medical clinic with a prenatal caregiver such as an OBGYN. Other implementations of module 310 generate a PND risk score as part of a user’s postnatal care.

[0071] Some implementations of module 310 report the PND risk score to a user, optionally along with additional information about PND or along with an intervention described elsewhere herein. With the user’s consent, some implementations of module 310 communicate a PND risk score to a user’s care provider (for example, primary care provider or OBGYN) via one or several channels, such as secure email. The user’s care provider has the option to refer the user for PND prevention or mitigation care, for example from the user’s existing care provider or another care provider (e.g., a psychologist or other mental health specialist). Some implementations of module 310 report the PND risk score along with an explanation of how the PND risk score was generated, to a user, the user’s care giver, or to another application (e.g., for model validation, additional research, or additional model training). Some non-limiting examples of an explanation are names, descriptions, and values of types of data that contributed most, or contributed at all, to generation of the user’s PND risk score.

[0072] Some implementations of module 310 are accessible via a clinic user authorization flow in which a clinic user (i.e., a caregiver of a user) receives a link to a user results management website, a clinic user accesses the user results management website which triggers a redirection to a single sign-on (SSO) website, a clinic user logs into the SSO website, a clinic user’s registered email address or other contact method receives a confirmation code, a cloud-based directory and identity management service such as Azure AD triggers an authorization chain, a clinic user logs in to the user results management website with the option to reduce secure sign on for future log-ins, and the clinic user is able to access CRUID related operations on data, view a dashboard of user results, and uploadDocket No. 124763-0015 additional data. Some implementations of module 310 are accessible via a user authorization flow in which a user’s contact method (e.g., email or phone) are derived from clinic data or a form the user completes, the user receives a link to a website for login and registration, the user’s registered email address or other registered contact method receives a confirmation code, a cloud-based directory and identity management service such as Azure AD triggers an authorization chain, the user logs in to a website or application implementing screening and intervention functionality with the option to reduce secure sign on for future log-ins, and the user is able to access a website or application implementing screening and intervention functionality.

[0073] Intervention module 320 determines that the PND risk score of the user is above a threshold score. Responsive to determining that the PND risk score of the user is above the threshold score, module 320 generates, using a trained conversational intervention agent executing on a mobile device of a user, an intervention intended to reduce the user’s PND risk score. In another implementation of module 320, the trained conversational intervention agent executes on another device, such as a website or an application executing on a server in a cloud environment.

[0074] Some implementations of module 320 offer a preventive treatment to the user. In some implementations of module 320, treatment is initiated, for example, before or during pregnancy (e.g., during the first pregnancy trimester) to reduce the risk of PND, in some estimates up to 50% or higher, or to reduce PND’s severity or to reduce PND’s severity and may continue after delivery for up to 12 months postpartum. In some implementations of module 320, treatment is initiated after the pregnancy ends. In some implementations of module 320, the treatment is based on the clinically validated ROSE protocol (Reach Out, Stay Strong, Essentials for mothers of newborns). With user consent, some implementations of module 320 offer an intervention, such as CBT or interpersonal, or other psychological support, to the partner of the pregnant person, also referred to as the significant other, or one or more family members engaged in the shared pregnancy experience and care before, during and after delivery.

[0075] In implementations of module 320, the intervention includes using a virtual conversational PND prevention agent to solicit a response from the user. One implementation of module 320 provides an intervention using a scripted chatbot implementation (a presently available technique), for example based on the ROSE clinical protocol. Another implementation of module 320 provides an intervention using a largeDocket No. 124763-0015 language model (LLM) that is designed, e.g., via fine-tuning, retrieval augmented generation, prompt engineering or other presently available techniques constraining the performance and behavior to a desired, highly scripted, professional yet conversational and engaging manner. Providing the intervention using a virtual conversational perinatal depression prevention agent or LLM enables a user to access the intervention in a selected language, such as English, Spanish, or Cantonese, and allows for rapid implementation of additional languages. One implementations of module 320 also adapts a level of language used in the intervention based on the user’s stated education or the language of that education (e.g., obtained from an intake form or the user’s electronic medical records). For example, a user with only an elementary school education or receiving an intervention in a language the user is less proficient in might be offered a lower level of language (e.g., simplified vocabulary and sentence structure) as compared to a user with a graduate degree receiving an intervention in their native language.

[0076] In implementations of module 320, the agent includes effective therapist characteristics, communication styles, the needs of specific cultural groups, and empathy. In implementations of module 320, performance of the agent is personalized which defines the individual user experience in terms of aspects such as duration and frequency of each session as well as other forms of gamification that drive optimal individual adherence to the virtual psychological support experience, during and after pregnancy and if PND does occur, during the PND to help adhering and completing the treatment. Some non-limiting examples of intervention session times are two minutes, three minutes, and five minutes. Some non-limiting examples of the number of intervention sessions are ten days and 28 days. In addition, some users complete interventions in fewer or more sessions than originally scheduled. Other session lengths and numbers of intervention sessions are also possible.

[0077] In some implementations of module 320, a session includes one or more of a checkin questionnaire (e.g., to be completed daily), topic-based chatbot content, and meditation exercises. In some implementations of module 320, the check-in questionnaire includes questions from PHQ2, PHQ8, or EDPS, presently available patient health questionnaires.

[0078] One non-limiting example of topic-based chatbot content is, “Did your mother ever share with you what it was like the first couple of months of having you as a baby?” with an option of selecting yes or no. If the user selects yes, an embodiment might respond, “That’s great! It’s helpful to learn from her experiences.” If the user selects no, an embodimentDocket No. 124763-0015 might respond, “That’s okay. Many women don’t talk about that time because it can be so personal or challenging. While some women find the early months of motherhood easier, many experience challenges during the first three months after birth. I want to share some honest realities with you — not to worry you, but to help you feel prepared and avoid feeling guilty if things feel tough at times. Knowing what to expect can make the experience less overwhelming and help you navigate these changes with more confidence.”

[0079] Another non-limiting example of topic-based chatbot content is, “Let’s think about what might change when baby arrives. Sometimes we focus so much on what we're giving up that we forget about what we're gaining. What do you think you might miss about your current life?” If the user responds, “Free time with my husband, losing my pre-pregnancy body, not easy to go out with friends without considering babysitting, less freedom with money,” module 320 might respond, “That’s a thoughtful way of reflecting on the changes ahead. It’s totally normal to miss some of the things you enjoyed before, like the freedom to spend time with your husband, your body, or being able to go out without extra planning.”

[0080] In some implementations of module 320, an intervention includes an opportunity to create a personalized virtual environment, such as a garden world and gardening activities, where the user picks their child’s birth flower, which the user will “nurture” over time from a seedling to a blossoming plant by completing small daily tasks. This approach aligns with the values of care, patience, and gentle growth, matching the self-compassionate progress module 320 is intended to help support.

[0081] Some implementations of module 320 generate an intervention using data from one or more sensors. For example, module 320 might use a presently available voice analysis technique to measure, from a sample of the user’s speech, biomarkers indicating that the user is depressed or stressed, and tailor an intervention accordingly. For example, if the user is particularly stressed, an intervention might include, “You seem particularly stressed today, would you like me to walk you through a breathing exercise that many folks find calming?” As another example, module 320 might use one or more biometric sensors to measure a user’s sleep time or sleep quality and offer an intervention such as, “I see you only slept four hours last night, would you like some suggestions to help you get more sleep?” As another example, module 320 might use one or more biometric sensors (e.g., an accelerometer, a heart rate sensor, a step counter, and the like) to measure a user’s level of physical activity and offer an intervention such as, “I see you got in a walk every day this week - great job!”Docket No. 124763-0015

[0082] In some implementations of module 320, to further personalize and gamify the experience with the goal of best possible adherence to session completion, upon consent from the user and other participants, an individual social network is created for the user leveraging the initial personal information provided by the user. This network can be drawn upon via publicly available social network data sources. The goal is to use the power of social networks to provide emotional and active support for the pregnant person during their journey, and thus in some embodiments an intervention includes use of the generated social network. For example, users might view each other's progress in a visually appealing and non-intrusive manner through a cohesive visual representation, such as growing flowers, helping to cultivate a sense of unity and reduce feelings of isolation. As another example, users might (with consent) interact with one another (e.g., by "watering" each other's flowers), serving as a simple yet impactful gesture of encouragement, a subtle form of social interaction designed to foster a sense of camaraderie and strengthen the bonds within the community of expectant mothers. As another example, module 320 might suggest an intervention a user’s social media contact might perform with the user. At any time during any session, the provider or an emergency hotline, e.g., to prevent harm to the user or others, can be contacted, for example, via audio, video or both or via messaging. In one embodiment, the provider or the emergency hotline may be contacted automatically based on the content of the conversation to preemptively reduce the risk of harm to self or others.

[0083] One implementation of module 320 includes one or more adherence protocols to facilitate an attrition-free completion of the psychotherapy or pharmacotherapy for those patients who do go on to develop PND requiring such treatment. These protocols leverage the disclosed individualized gamification features to enable an adherence protocol. Some embodiments monitor user engagement with one or more interventions. If a user does not participate in one or more interventions, does not complete one or more interventions, or an embodiment determines from the user’s answers or other user data that user engagement is below a threshold level, embodiments trigger execution of an adherence protocol. Some implementations of module 320, instead of or in addition to an adherence protocol, engage the user via proactive nudges via email, text message, or using another communication method.

[0084] Some implementations of module 320, during or after completion of a PND intervention protocol, generate an updated PND risk score of the user, using the trained PND screening model. If the updated PND risk score is above a threshold, or is trendingDocket No. 124763-0015 upwards, some implementations of module 320 adjust the PND intervention protocol, and other implementations of module 320 (with user consent) alert a caregiver, partner, or other family member to the updated PND risk score or trend.

[0085] Application 222 includes a postpartum preventive wrap-up session. In one implementation of application 222, this session is based on the ROSE protocol. In some implementations of application 222, the partner and / or the family members are also involved in the sessions prepared specifically for these family members.

[0086] FIG. 4 depicts an example of components of a product implementing mobile device-based prenatal screening and intervention for perinatal depression, in accordance with an illustrative embodiment. Screening module 310 and intervention module 320 are the same as screening module 310 and intervention module 320 in FIG. 3. The example can be executed using application 222 in FIG. 2.

[0087] Product 400 includes functionality of screening module 310 and intervention module 320. Module 310 generates, using a trained perinatal depression (PND) screening model executing on a mobile device of a user and, among other data, an intake survey performed at a user’s 22- week appointment with a care provider, a PND risk score of the user. Module 320 implements a sentiment analysis and a virtual conversational perinatal depression prevention agent to solicit a response from the user. At any time during any session, the provider or an emergency hotline, e.g., to prevent harm to the user or others, can be contacted, for example, via audio, video or both or via messaging. In one implementation of application 222, the provider or the emergency hotline may be contacted automatically based on the content of the conversation to preemptively reduce the risk of harm to self or others.

[0088] FIG. 5 depicts an example of authentication flows used in mobile device-based prenatal screening and intervention for perinatal depression, in accordance with an illustrative embodiment.

[0089] Some implementations of application 222 are accessible via a clinic user authorization flow, depicted in authentication 510, in which a clinic user (i.e., a caregiver of a user) receives a link to a user results management website, a clinic user accesses the user results management website which triggers a redirection to a single sign-on (SSO) website, a clinic user logs into the SSO website, a clinic user’s registered email address or other contact method receives a confirmation code, a cloud-based directory and identity management service such as Azure AD triggers an authorization chain, a clinic user logs inDocket No. 124763-0015 to the user results management website with the option to reduce secure sign on for future log-ins, and the clinic user is able to access CRUID related operations on data, view a dashboard of user results, and upload additional data. Some implementations of application 222 are accessible via a user authorization flow, depicted in authentication 520, in which a user’s contact method (e.g., email or phone) are derived from clinic data or a form the user completes, the user receives a link to a website for login and registration, the user’s registered email address or other registered contact method receives a confirmation code, a cloud-based directory and identity management service such as Azure AD triggers an authorization chain, the user logs in to a website or application implementing screening and intervention functionality with the option to reduce secure sign on for future log-ins, and the user is able to access a website or application implementing screening and intervention functionality.

[0090] FIG. 6 depicts an example of a product architecture implementing mobile devicebased prenatal screening and intervention for perinatal depression, in accordance with an illustrative embodiment. In particular, product architecture 610 depicts a product architecture in which authentications 510 and 520 in FIG. 5 can be implemented.

[0091] FIG. 7 depicts a flowchart of an example process for mobile device-based prenatal screening and intervention for perinatal depression, in accordance with an illustrative embodiment. Process 700 can be implemented in application 222 in FIG. 2.

[0092] At block 702, the process generates, using a trained perinatal depression (PND) screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data of the user, a PND risk score of the user. At block 704, the process determines that the PND risk score of the user is above a threshold score. At block 706, the process generates, using a trained conversational intervention agent executing on a mobile device of a user, responsive to determining that the PND risk score of the user is above the threshold score, an intervention intended to reduce the PND risk score of the user. Then the process ends.

[0093] Many of the above-described features and applications may be implemented as software processes that are specified as a set of instructions recorded on a computer- readable storage medium (alternatively referred to as computer-readable media, machine- readable media, or machine-readable storage media). When these instructions are executed by one or more processing unit(s) (e.g., one or more processors, cores of processors, or other processing units), they cause the processing unit(s) to perform the actions indicated inDocket No. 124763-0015 the instructions. Examples of computer-readable media include, but are not limited to, RAM, ROM, read-only compact discs (CD-ROM), recordable compact discs (CD-R), rewritable compact discs (CD-RW), read-only digital versatile discs (e.g., DVD-ROM, duallayer DVD-ROM), a variety of recordable / rewritable DVDs (e.g., DVD-RAM, DVD-RW, DVD+RW, etc.), flash memory (e.g., SD cards, mini-SD cards, micro-SD cards, etc.), magnetic and / or solid state hard drives, ultra-density optical discs, any other optical or magnetic media, and floppy disks. In one or more embodiments, the computer- readable media does not include carrier waves and electronic signals passing wirelessly or over wired connections, or any other ephemeral signals. For example, the computer- readable media may be entirely restricted to tangible, physical objects that store information in a form that is readable by a computer. In one or more embodiments, the computer-readable media is non-transitory computer-readable media, computer-readable storage media, or non- transitory computer-readable storage media.

[0094] In one or more embodiments, a computer program product (also known as a program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, declarative or procedural languages, and it can be deployed in any form, including as a standalone program or as a module, component, subroutine, object, or other unit suitable for use in a computing environment. A computer program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.

[0095] While the above discussion primarily refers to microprocessor or multi-core processors that execute software, one or more embodiments are performed by one or more integrated circuits, such as application specific integrated circuits (ASICs) or field programmable gate arrays (FPGAs). In one or more embodiments, such integrated circuits execute instructions that are stored on the circuit itself.

[0096] While this specification contains many specifics, these should not be construed as limitations on the scope of what may be claimed, but rather as descriptions of particular implementations of the subject matter. Certain features that are described in thisDocket No. 124763-0015 specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a subcombination.

[0097] Those of skill in the art would appreciate that the various illustrative blocks, modules, elements, components, methods, and algorithms described herein may be implemented as electronic hardware, computer software, or combinations of both. To illustrate this interchangeability of hardware and software, various illustrative blocks, modules, elements, components, methods, and algorithms have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application. Various components and blocks may be arranged differently (e.g., arranged in a different order, or partitioned in a different way), all without departing from the scope of the subject technology.

[0098] It is understood that any specific order or hierarchy of blocks in the processes disclosed is an illustration of example approaches. Based upon implementation preferences, it is understood that the specific order or hierarchy of blocks in the processes may be rearranged, or that not all illustrated blocks be performed. Any of the blocks may be performed simultaneously. In one or more embodiments, multitasking and parallel processing may be advantageous. Moreover, the separation of various system components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

[0099] The subject technology is illustrated, for example, according to various aspects described above. The present disclosure is provided to enable any person skilled in the art to practice the various aspects described herein. The disclosure provides various examples of the subject technology, and the subject technology is not limited to these examples. VariousDocket No. 124763-0015 modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects.

[0100] A reference to an element in the singular is not intended to mean “one and only one” unless specifically stated, but rather “one or more.” Unless specifically stated otherwise, the term “some” refers to one or more. Pronouns in the masculine (e.g., his) include the feminine and neuter gender (e.g., her and its) and vice versa. Headings and subheadings, if any, are used for convenience only and do not limit the disclosure.

[0101] To the extent that the terms “include,” “have,” or the like is used in the description or the claims or clauses, such term is intended to be inclusive in a manner similar to the term “comprise” as “comprise” is interpreted when employed as a transitional word in a claim.

[0102] The word “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments. In one aspect, various alternative configurations and operations described herein may be considered to be at least equivalent.

[0103] As used herein, the phrase “at least one of' preceding a series of items, with the terms “and” or “or” to separate any of the items, modifies the list as a whole, rather than each member of the list (i.e., each item). The phrase “at least one of’ does not require selection of at least one item; rather, the phrase allows a meaning that includes at least one of any one of the items, and / or at least one of any combination of the items, and / or at least one of each of the items. By way of example, the phrases “at least one of A, B, and C” or “at least one of A, B, or C” each refer to only A, only B, or only C; any combination of A, B, and C; and / or at least one of each of A, B, and C.

[0104] A phrase such as an “aspect” does not imply that such aspect is essential to the subject technology or that such aspect applies to all configurations of the subject technology. A disclosure relating to an aspect may apply to all configurations, or one or more configurations. An aspect may provide one or more examples. A phrase such as an aspect may refer to one or more aspects and vice versa. A phrase such as an “embodiment” does not imply that such embodiment is essential to the subject technology or that such embodiment applies to all configurations of the subject technology. A disclosure relating to an embodiment may apply to all embodiments, or one or more embodiments. An embodiment may provide one or more examples. A phrase such as an embodiment may referDocket No. 124763-0015 to one or more embodiments and vice versa. A phrase such as a “configuration” does not imply that such configuration is essential to the subject technology or that such configuration applies to all configurations of the subject technology. A disclosure relating to a configuration may apply to all configurations, or one or more configurations. A configuration may provide one or more examples. A phrase such as a configuration may refer to one or more configurations and vice versa.

[0105] In one aspect, unless otherwise stated, all measurements, values, ratings, positions, magnitudes, sizes, and other specifications that are set forth in this specification, including in the claims or clauses that follow, are approximate, not exact. In one aspect, they are intended to have a reasonable range that is consistent with the functions to which they relate and with what is customary in the art to which they pertain. It is understood that some or all steps, operations, or processes may be performed automatically, without the intervention of a user.

[0106] Method claims or clauses may be provided to present elements of the various steps, operations, or processes in a sample order, and are not meant to be limited to the specific order or hierarchy presented.

[0107] In one aspect, a method may be an operation, an instruction, or a function and vice versa. In one aspect, a claim may be amended to include some or all of the words (e.g., instructions, operations, functions, or components) recited in other one or more claims, one or more words, one or more sentences, one or more phrases, one or more paragraphs, and / or one or more claims.

[0108] All structural and functional equivalents to the elements of the various configurations described throughout this disclosure that are known or later come to be known to those of ordinary skill in the art are expressly incorporated herein by reference and intended to be encompassed by the subject technology. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the above description. No claim element is to be construed under the provisions of 35 U.S.C. §112, sixth paragraph, unless the element is expressly recited using the phrase “means for” or, in the case of a method claim, the element is recited using the phrase “step for.”

[0109] The Title, Background, and Brief Description of the Drawings of the disclosure are hereby incorporated into the disclosure and are provided as illustrative examples of the disclosure, not as restrictive descriptions. It is submitted with the understanding that theyDocket No. 124763-0015 will not be used to limit the scope or meaning of the claims. In addition, in the Detailed Description, it can be seen that the description provides illustrative examples, and the various features are grouped together in various embodiments for the purpose of streamlining the disclosure. This method of disclosure is not to be interpreted as reflecting an intention that the included subject matter requires more features than are expressly recited in any claim. Rather, as the claims reflect, inventive subject matter lies in less than all features of a single disclosed configuration or operation. The claims are hereby incorporated into the Detailed Description, with each claim standing on its own to represent separately patentable subject matter.

[0110] The claims or clauses are not intended to be limited to the aspects described herein but are to be accorded the full scope consistent with the language of the claims and to encompass all legal equivalents. Notwithstanding, none of the claims are intended to embrace subject matter that fails to satisfy the requirement of 35 U.S.C. § 101, 102, or 103, nor should they be interpreted in such a way.

[0111] Embodiments consistent with the present disclosure may be combined with any combination of features or aspects of embodiments described herein.

Claims

Docket No. 124763-0015CLAIMS1. A computer-implemented method comprising: generating, using a trained perinatal depression (PND) screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data of the user, a PND risk score of the user; determining that the PND risk score of the user is above a threshold score; and generating, using a trained conversational intervention agent executing on the mobile device of the user, responsive to determining that the PND risk score of the user is above the threshold score, an intervention intended to reduce the PND risk score of the user.

2. The computer-implemented method of claim 1, wherein generating the PND risk score is performed while the user is pregnant.

3. The computer-implemented method of claim 1, further comprising: communicating, to a care provider of the user, the PND risk score and an explanation of values of types of data contributing to the PND risk score.

4. The computer-implemented method of claim 1 , wherein the intervention is generated using data from a biometric sensor.

5. The computer- implemented method of claim 1, wherein the intervention comprises use of an individual social network generated for the user.

6. The computer-implemented method of claim 1 , wherein the intervention comprises using a virtual conversational perinatal depression prevention agent to solicit a response from the user.

7. The computer-implemented method of claim 6, wherein the virtual conversational perinatal depression prevention agent is implemented using a scripted chatbot based on a clinical protocol.

8. The computer-implemented method of claim 6, wherein the virtual conversational perinatal depression prevention agent is implemented using a large language model.Docket No. 124763-00159. The computer-implemented method of claim 6, wherein the virtual conversational perinatal depression prevention agent uses a level of language based on an education level of the user.

10. A non-transitory computer-readable medium storing a program, which when executed by a computer, configures the computer to: generate, using a trained perinatal depression (PND) screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data of the user, a PND risk score of the user; determine that the PND risk score of the user is above a threshold score; and generate, using a trained conversational intervention agent executing on the mobile device of the user, responsive to determining that the PND risk score of the user is above the threshold score, an intervention intended to reduce the PND risk score of the user.

11. The non-transitory computer-readable medium of claim 10, wherein generating the PND risk score is performed while the user is pregnant.

12. The non-transitory computer-readable medium of claim 10, wherein the program, when executed by the computer, further configures the computer to: communicate, to a care provider of the user, the PND risk score and an explanation of values of types of data contributing to the PND risk score.

13. The non-transitory computer- readable medium of claim 10, wherein the intervention is generated using data from a biometric sensor.

14. The non-transitory computer-readable medium of claim 10, wherein the intervention comprises use of an individual social network generated for the user.

15. The non-transitory computer- readable medium of claim 10, wherein the intervention comprises using a virtual conversational perinatal depression prevention agent to solicit a response from the user.Docket No. 124763-001516. The non-transitory computer- readable medium of claim 15, wherein the virtual conversational perinatal depression prevention agent is implemented using a scripted chatbot based on a clinical protocol.

17. The non-transitory computer-readable medium of claim 15, wherein the virtual conversational perinatal depression prevention agent is implemented using a large language model.

18. The non-transitory computer-readable medium of claim 15, wherein the virtual conversational perinatal depression prevention agent uses a level of language based on an education level of the user.

19. A system comprising: a processor; and a non-transitory computer readable medium storing a set of instructions, which when executed by the processor, configure the system to: generate, using a trained perinatal depression (PND) screening model executing on a mobile device of a user and health history data, demographic data, and biosensor data of the user, a PND risk score of the user; determine that the PND risk score of the user is above a threshold score; and generate, using a trained conversational intervention agent executing on the mobile device of the user, responsive to determining that the PND risk score of the user is above the threshold score, an intervention intended to reduce the PND risk score of the user.

20. The system of claim 19, wherein generating the PND risk score is performed while the user is pregnant.

Images