A system for detecting micronutrient deficiencies using facial markers

Abstract

A system (100) for detecting micronutrient deficiencies using facial markers, comprising: a data input module (102) configured to receive digital facial images; an image processing unit (104) configured to process the input digital facial images; a central processing unit (106) configured to detect micronutrient deficiencies based on the input digital facial image; a recommendation engine (108) configured to generate suggestions for vitamin-rich foods; and a user interface module (110) that includes a display (112) configured to show detection results and recommendations.

Description

FIELD OF THE INVENTION

[0001] The present disclosure relates to a system for detecting micronutrient deficiencies using facial markers. More specifically, the present invention relates to a non-invasive system for identifying potential deficiencies by analyzing facial features captured by digital images. The system integrates deep learning models and advanced image processing techniques to predict and map vitamin deficiencies and associated skin conditions. BACKGROUND OF THE INVENTION

[0002] Micronutrient deficiencies are a significant global health problem and lead to various health issues, including skin problems, weakened immune systems, and chronic diseases. Current diagnostic methods often require invasive procedures or are not easily accessible to all populations. This invention addresses the need for an accessible, rapid, and non-invasive method for detecting micronutrient deficiencies. By utilizing facial markers, the system can analyze visual cues to identify potential deficiencies, providing a practical solution for early intervention and improved health outcomes. SUMMARY OF THE INVENTION

[0003] This disclosure relates to a micronutrient deficiency detection system using facial markers. The proposed micronutrient deficiency detection system uses facial markers to identify potential deficiencies. The system uses YOLOv5 and custom models enhanced with attention mechanisms to analyze collected facial data and associate it with potential vitamin deficiencies and associated skin problems. This approach leverages existing deep learning techniques tailored to facial analysis to create an accessible tool for healthcare providers and individuals seeking to monitor their micronutrient levels.

[0004] The present disclosure is intended to provide a system for detecting micronutrient deficiencies using facial landmarks. The system comprises: a data input module configured to receive digital facial images; an image processing unit configured to process the input digital facial images; a central processing unit configured to detect micronutrient deficiencies based on the input digital facial image; a recommendation engine configured to generate suggestions for vitamin-rich foods; and a user interface module including a display configured to display detection results and recommendations.

[0005] One objective of the present disclosure is to provide a system for detecting micronutrient deficiencies using facial markers.

[0006] Another objective of the present disclosure is to develop a non-invasive system for detecting vitamin deficiency through AI-assisted facial feature analysis.

[0007] Another objective of the present disclosure is to provide an easily accessible tool for associating facial skin disorders with specific vitamin deficiencies.

[0008] Another objective of this disclosure is to provide personalized nutritional recommendations based on identified deficiencies.

[0009] To further clarify the advantages and features of the present disclosure, a more detailed description of the invention will be given with reference to specific embodiments thereof illustrated in the accompanying drawings. It should be noted that these drawings represent only typical embodiments of the invention and are therefore not to be considered limiting its scope. The invention will be described and explained in additional detail and in greater detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE CHARACTERS

[0010] These and other features, aspects, and advantages of the present disclosure will be better understood when the following detailed description is read with reference to the accompanying drawings, in which like characters represent like parts throughout. Fig. 1 shows a block diagram of a system for detecting micronutrient deficiencies using facial landmarks according to an embodiment of the present disclosure; and Fig. 2 illustrates a facial micronutrient deficiency system according to an embodiment of the present disclosure.

[0011] Furthermore, those skilled in the art will appreciate that elements in the drawings are shown for convenience and may not necessarily be drawn to scale. For example, the flowcharts illustrate the method by key steps to enhance understanding of aspects of the present disclosure. Moreover, with respect to device construction, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details relevant to understanding embodiments of the present disclosure in order not to clutter the drawings with details that would be readily apparent to those skilled in the art who would benefit from the description herein. DETAILED DESCRIPTION:

[0012] To facilitate an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and described in specific language. It is to be understood, however, that this is not intended to limit the scope of the invention, as such changes and further modifications to the illustrated system and such further applications of the principles of the invention as illustrated therein are contemplated as would normally occur to one skilled in the art to which the invention pertains.

[0013] It will be understood by one skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not intended to be limiting thereof.

[0014] References in this specification to "one aspect," "another aspect," or similar language mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Therefore, occurrences of the phrase "in one embodiment," "in another embodiment," and similar language throughout this specification may or may not all refer to the same embodiment.

[0015] The terms "comprises," "comprising," or other variations thereof are intended to cover non-exclusive inclusion, such that a process or method comprising a list of steps not only includes those steps, but may also include other steps not expressly listed or inherent in such process or method. Likewise, one or more devices or subsystems or elements or structures or components preceded by "comprises...a" does not preclude, without further limitation, the existence of other devices or other subsystems or other elements or other structures or other components or additional devices or additional subsystems or additional elements or additional structures or additional components.

[0016] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The system, methods, and examples provided herein are for illustrative purposes only and should not be considered limiting.

[0017] Embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0018] The functional units described in this specification have been referred to as devices. A device may be implemented in programmable hardware devices such as processors, digital signal processors, central processing units, field-programmable gate arrays, programmable array logic, programmable logic devices, cloud processing systems, or the like. The devices may also be implemented in software for execution by various types of processors. An identified device may contain executable code and may, for example, comprise one or more physical or logical blocks of computer instructions, which may be organized, for example, as an object, procedure, function, or other construct.However, the executable file of an identified device need not be physically located together, but may comprise different instructions stored in different locations which, when logically linked together, constitute the device and fulfill the stated purpose of the device.

[0019] Indeed, executable code of a device or module may be a single instruction or multiple instructions, and may even be distributed across several different code segments, among different applications, and across multiple storage devices. Similarly, operational data may be identified and represented within the device and embodied in any suitable form and organized in any suitable type of data structure. The operational data may be captured as a single set of data or distributed across different locations, including different storage devices, and may exist, at least in part, as electronic signals in a system or network.

[0020] References in this specification to "a selected embodiment," "an embodiment," or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the disclosed subject matter. Therefore, the appearances of the phrases "a selected embodiment," "in an embodiment," or "in an embodiment" in various places in this specification do not necessarily refer to the same embodiment.

[0021] Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to provide a thorough understanding of embodiments of the disclosed subject matter. However, one of ordinary skill in the art will recognize that the disclosed subject matter may be practiced without one or more of the specific details, or with different methods, components, materials, etc. In other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of the disclosed subject matter.

[0022] According to the example embodiments, the disclosed computer programs or modules may be executed in many example ways, for example, as an application stored in the memory of a device or as a hosted application running on a server and communicating with the device application or browser using a variety of standard protocols such as TCP / IP, HTTP, XML, SOAP, REST, JSON, and other suitable protocols. The disclosed computer programs may be written in example programming languages ​​that execute from memory on the device or from a hosted server, for example, BASIC, COBOL, C, C++, Java, Pascal, or scripting languages ​​such as JavaScript, Python, Ruby, PHP, Perl, or other suitable programming languages.

[0023] Some of the disclosed embodiments comprise or otherwise involve data transmission over a network, such as communicating various inputs or files over the network. The network may comprise, for example, one or more of the following: the Internet, wide area networks (WANs), local area networks (LANs), analog or digital wired and wireless telephone networks (e.g., a PSTN, Integrated Services Digital Network (ISDN), a cellular network, and Digital Subscriber Line (xDSL)), radio, television, cable, satellite, and / or any other delivery or tunneling mechanism for transmitting data. The network may comprise multiple networks or subnetworks, each of which may comprise, for example, a wired or wireless data path. The network may comprise a circuit-switched voice network, a packet-switched data network, or any other network capable of transmitting electronic communications.For example, the network may include networks based on Internet Protocol (IP) or Asynchronous Transfer Mode (ATM), and may support voice using, for example, VoIP, Voice over ATM, or other comparable protocols used for voice data communications. In one implementation, the network includes a cellular network configured for the exchange of text or SMS messages.

[0024] Examples of the network include, but are not limited to, a Personal Area Network (PAN), a Storage Area Network (SAN), a Home Area Network (HAN), a Campus Area Network (CAN), a Local Area Network (LAN), a Wide Area Network (WAN), a Metropolitan Area Network (MAN), a Virtual Private Network (VPN), an Enterprise Private Network (EPN), the Internet, a Global Area Network (GAN), and so on.

[0025] Fig. 1 shows a block diagram of a system (100) for detecting micronutrient deficiencies using facial landmarks according to an embodiment of the present disclosure.

[0026] With reference to Fig. 1, the system (100) comprises a data input module (102) configured to receive digital facial images; an image processing unit (104) configured to process the input digital facial images; a central processing unit (106) configured to detect micronutrient deficiencies based on the input digital facial image; a recommendation engine (108) configured to generate suggestions for vitamin-rich foods; and a user interface module (110) having a display (112) configured to display detection results and recommendations.

[0027] In one embodiment, the data input module (102) is further configured to validate image quality parameters, standardize image dimensions, perform preliminary facial feature extraction, and prepare images for model processing.

[0028] In one embodiment, the image processing unit (104) comprises: a primary recognition model (104a) implementing the YOLOv5 architecture and configured to analyze facial features, wherein the primary recognition model (104a) is further configured to detect multiple skin conditions, including acne, blackheads, dark circles, dark spots, eczema, freckles, nodules, skin redness, and wrinkles, to generate confidence scores for detected conditions, and to identify spatial positions of detected conditions;and a secondary recognition model (104b) implementing YOLOv5+ with a Coordinate Attention module (104c) and configured to improve recognition accuracy, wherein the secondary recognition model (104b) with Coordinate Attention module (104c) is further configured to maintain position information during feature recognition, improve recognition accuracy through attention mechanisms, generate position-related feature maps, and improve recognition precision for overlapping states.;

[0029] In one embodiment, the centralized processing unit (106) comprises: a model selection processor (106a) configured to switch between recognition models; an inference engine (106b) configured to process analyzed facial data; a deficiency assignment module (106c) configured to correlate recognized facial features with vitamin deficiencies; and a data management module (106d) configured to manage historical recognition records, track recognition accuracy metrics, store user dietary preferences, and manage system configuration parameters.

[0030] In one embodiment, the model selection processor (106a) is further configured to enable concurrent processing by both models, compare detection results between models, select optimal results based on confidence scores, and maintain model performance metrics.

[0031] In one embodiment, the inference engine (106b) is further configured to process detection results in real time, aggregate insights from both detection models, generate comprehensive analysis reports, and maintain detection history for comparative analysis.

[0032] In one embodiment, the deficiency assignment module (106c) is further configured to maintain a database of correlations between skin condition and vitamin deficiency, calculate deficiency probability scores, create deficiency severity assessments, and track deficiency patterns over time.

[0033] In one embodiment, the recommendation engine (108) is further configured to generate personalized dietary recommendations, prioritize food suggestions based on the severity of the deficiency, provide alternative food options, and update recommendations based on user feedback.

[0034] In one embodiment, the user interface module (110) with the display (112) is further configured to display detected skin conditions with spatial markers, visualize defect mapping results, display comparative model recognition results, and enable data entry of digital facial images and model selection preferences.

[0035] In one embodiment, the data input module (102), the image processing unit (104), the centralized processing unit (106), the recommendation engine (108), and then the user interface module (110) may be implemented in programmable hardware devices such as processors, digital signal processors, central processing units, field-programmable gate arrays, programmable array logic, programmable logic devices, cloud processing systems, or the like.

[0036] The present invention relates to a system for detecting micronutrient deficiencies based on facial markers. The system integrates deep learning models and advanced image processing techniques to predict and map vitamin deficiencies and associated skin diseases.

[0037] Fig. 2 illustrates a facial micronutrient deficiency system according to an embodiment of the present disclosure.

[0038] Fig. As shown in Figure 2, the system uses a deep learning model along with a modified model that integrates attention mechanisms to process facial data and identify potential vitamin deficiencies and associated skin problems. This system applies advanced deep learning techniques specifically for facial analysis, providing an accessible solution for healthcare professionals and individuals who want to track their micronutrient levels.

[0039] The invention aims to help users identify their vitamin deficiencies by analyzing facial skin problems. Vitamins are vital organic compounds necessary for growth and proper functioning, and their deficiencies can manifest as visible skin problems on the face. This system associates detected skin problems with specific vitamin deficiencies and offers nutritional recommendations to encourage users to eat a healthy, vitamin-rich diet.

[0040] The architecture consists of several core components: model pre-training, YOLOv5 model, YOLOv5+ integrated with a Coordinate Attention (CA) module, and backend and frontend systems.

[0041] The model is trained on datasets collected online, with a focus on skin conditions. Classification categories include acne, blackheads, dark circles, dark spots, eczema, freckles, healthy skin, nodules, redness, and wrinkles. Datasets are first filtered based on whether a face is identified using a pre-trained Haar cascade classifier and then included in the dataset. Augmentation was enabled with Roboflow, and a larger dataset was created. AutoOrient and Resize stretch to 640 x 640, random rotation between -15 degrees and +15 degrees, random saturation change between -25% and +25%, random brightness change between -20% and +20%, and YOLOv5 performs online augmentation during training.

[0042] Two models were trained independently: the standard YOLOv5 and a modified YOLOv5+ with the integrated Coordinate Attention module. The standard model achieved an accuracy of approximately 82%, while the modified model showed improved accuracy of approximately 85% and effectively identified the locations of skin problems. Coordinate Attention contributes to improving model accuracy by helping models more accurately locate objects of interest by increasing their "attention" and feature extraction. YOLOv5 is the benchmark against YOLOv5 + Coordinate Attention, and YOLOv5 + Coordinate Attention provides higher accuracy compared to YOLOv5. A website system that has the choice to use both separately takes advantage of both models and classes that may not receive sufficient focus in a single model.

[0043] YOLOv5 is a state-of-the-art object detection model specifically trained to detect skin conditions in this application. The model's performance was improved by incorporating the Coordinate Attention module, a specialized channel attention mechanism that retains positional information critical for accurate detection. The model was trained on a dataset covering 10 different skin conditions, including acne, blackheads, dark circles, dark spots, eczema, freckles, healthy skin, nodules, skin redness, and wrinkles. Each of these classes has unique visual features that the model is designed to detect. For example, acne is characterized by red pustules or whiteheads, while eczema appears as rough, scaly patches. Wrinkles are characterized by lines or creases, often found around areas such as the eyes or mouth.This differentiation allows the model to accurately categorize based on facial features that are unique to each skin condition.

[0044] The enhanced version, YOLOv5+CA, with the Coordinate Attention (CA) module, has an accuracy of approximately 85%, slightly higher than YOLOv5's 82% accuracy. The CA module helps preserve spatial detail and can improve detection for a specific class, which is beneficial when specific skin problems require targeted accuracy. However, if users are seeking robust predictions for multiple skin problem classes rather than targeted improvements for specific classes, the standard YOLOv5 model may be more suitable. YOLOv5 balances accuracy across all classes, making it a versatile choice for detecting multiple conditions without a strong emphasis on any one condition. The proposed system suggests potential nutrient deficiencies corresponding to the detected skin problems.

[0045] The backend, developed with Python Flask, processes the images submitted by users, sends them to the model for analysis, and returns the inference results to the frontend. The system does not include a scoring system or a threshold to indicate the severity of a vitamin deficiency. The model detects the presence of skin conditions associated with specific nutrient deficiencies, but does not provide a measure of the severity of the deficiency.

[0046] The frontend, built with React.js and Tailwind CSS, allows users to upload images for analysis and select which model to use for inference. After uploading the images and selecting the model, the system provides the user with detailed results, including the identified skin problem (if any), the corresponding vitamin deficiency, and recommendations for vitamin-rich foods to correct the deficiency.

[0047] The model achieves 85% accuracy in detecting skin conditions using the YOLOv5+CA model. While the associations between skin conditions and micronutrient deficiencies are based on research-based relationships, there is no specific clinical validation or expert feedback for these associations. The system detects multiple vitamin deficiencies for each identified skin condition class, without prioritizing any one deficiency. This means that for each detected skin condition (e.g., acne or dark circles under the eyes), all associated vitamin deficiencies are equally represented, allowing users to understand the full range of potential deficiencies associated with each condition.

[0048] The system detects facial skin problems, matches them with corresponding vitamin deficiencies, and recommends nutrient-rich foods. These food suggestions are based on general recommendations for specific nutritional needs, without relying on nutrient density values ​​or specialized databases. The recommendations are comprehensive and list foods with significant nutrient content, leaving dietary restrictions, allergies, or regional preferences to the user's discretion.

[0049] Currently, the system does not have a user login feature, meaning all users can access the interface, upload images, and receive model predictions. However, user login is planned to save previous entries and recommendations. The system will also provide greater transparency by displaying the confidence level of model predictions. The severity of skin problems is currently neither rated nor displayed, but this is a planned future improvement.

[0050] The system has been tested locally and supports multiple user uploads, ensuring anonymity. Submitted images are temporarily stored for processing and deleted upon leaving the system. In production, the system will include load balancing and a microservices architecture for handling user requests. Its responsive design ensures compatibility with multiple devices and provides a seamless user experience across mobile and desktop platforms.

[0051] Performance metrics such as precision, recall, and F1 score are reported alongside accuracy (82% for the baseline YOLOv5 model and 85% for the modified Coordinate Attention model). These metrics improve understanding of the system's effectiveness and compensate for false positives and negatives. The system was tested on various datasets with different lighting, skin tones, and resolutions to ensure robustness and adaptability to real-world scenarios.

[0052] The system is limited to detecting 10 specific skin conditions (acne, blackheads, dark circles, dark spots, eczema, freckles, healthy skin, nodules, redness, and wrinkles). It may not be able to accurately identify skin problems outside these categories or distinguish between similar conditions outside of its training data. Plans include expanding the range of detectable conditions and introducing severity detection for more personalized insights.

[0053] The system architecture, which leverages deep learning and attention mechanisms, supports a potential extension to more comprehensive facial analysis. Additional health indicators such as texture or tone changes could be integrated through retraining the model or expanded modules.

[0054] This non-invasive application is for guidance only and offers nutrition-based recommendations for skin concerns without the use of medical devices. It has not yet been validated or recommended by medical professionals.

[0055] The present invention relates to a system that identifies and localizes facial markers of micronutrient deficiency using computer vision. This invention develops a system for identifying facial markers of micronutrient deficiency. The system identifies and localizes the facial markers and recommends appropriate nutrient-rich foods.

[0056] The drawings and the foregoing description provide examples of embodiments. Those skilled in the art will recognize that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be separated into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, orders of the processes described herein may be changed and are not limited to the manner described herein. Furthermore, the actions of any flowchart need not be implemented in the order shown; nor do all actions necessarily need to be performed. Also, those actions that are not dependent on other actions may be performed in parallel with the other actions. The scope of the embodiments is in no way limited by these specific examples.Numerous variations, whether explicitly stated in the specification or not, such as differences in structure, dimensions, and material use, are possible. The scope of the embodiments is at least as broad as indicated in the following claims.

[0057] Advantages, other benefits, and solutions to problems have been described above with reference to specific embodiments. However, the advantages, benefits, solutions to problems, and any components that may cause an advantage or solution to occur or become more apparent are not to be construed as a critical, required, or essential feature or component of any or all of the claims. REFERENCES 100 A system for detecting micronutrient deficiencies using facial markers. 102 Data entry module 104 Image processing unit 104a Primary detection model 104b secondary detection model 104c Module “Coordinated Attention” 106 central processing unit 106a Model selection processor 106b Inference engine 106c Defect mapping module 106d Data management module 108 Recommendation Engine 110 User interface module 112 Advertisement 202 Upload 204 Model selection? 206 Model-1 208 Model 2 (Modified) 210 Trained model 212 Training model 214 Attention addition to model 216 Recorded data set 218 Location of the skin problem 220 Skin problem 222 Mapped vitamin deficiency 224 Output to Frontend 226 Conclusion 228 Frontend 230 backend 232 Model pre-training

Claims

[1] A system (100) for detecting micronutrient deficiencies using facial markers, comprising: a data input module (102) configured to receive digital facial images; an image processing unit (104) configured to process the input digital facial images; a central processing unit (106) configured to detect micronutrient deficiencies based on the input digital facial image; a recommendation engine (108) configured to generate suggestions for vitamin-rich foods; and a user interface module (110) comprising a display (112) configured to display detection results and recommendations. [2] The system (100) of claim 1, wherein the data input module is further configured to validate image quality parameters, standardize image dimensions, perform preliminary facial feature extraction, and prepare images for model processing. [3] System (100) according to claim 1, wherein the image processing unit (104) comprises: a primary recognition model (104a) implementing a YOLOv5 architecture and configured to analyze facial features, wherein the primary recognition model (104a) is further configured to detect multiple skin conditions, including acne, blackheads, dark circles, dark spots, eczema, freckles, nodules, skin redness, and wrinkles; to generate confidence scores for detected conditions; and to identify spatial locations of detected conditions; and a secondary recognition model (104b) implementing YOLOv5+ with a Coordinate Attention module (104c) configured to improve recognition accuracy, wherein the secondary recognition model (104b) with Coordinate Attention module (104c) is further configured to maintain position information during feature recognition, improve recognition accuracy through attention mechanisms, generate position-related feature maps, and improve recognition precision under overlap conditions. [4] The system (100) of claim 1, wherein the centralized processing unit (106) comprises: a model selection processor (106a) configured to switch between recognition models; an inference engine (106b) configured to process analyzed facial data; a deficiency mapping module (106c) configured to correlate detected facial features with vitamin deficiencies; and a data management module (106d) configured to manage historical detection records, track detection accuracy measurements, store user dietary preferences, and manage system configuration parameters. [5] The system (100) of claim 4, wherein the model selection processor (106a) is further configured to enable concurrent processing by both models, compare recognition results between models, select optimal results based on confidence scores, and maintain model performance metrics. [6] The system (100) of claim 4, wherein the inference engine (106b) is further configured to process recognition results in real time, summarize insights from both recognition models, generate comprehensive analysis reports, and maintain recognition history for comparative analysis. [7] The system (100) of claim 4, wherein the deficiency assignment module (106c) is further configured to maintain a database of correlations between skin condition and vitamin deficiency, calculate deficiency probability scores, create deficiency severity assessments, and track deficiency patterns over time. [8] The system (100) of claim 1, wherein the recommendation engine (108) is further configured to generate personalized nutritional recommendations, prioritize food suggestions based on the severity of the deficiency, provide alternative food options, and update recommendations based on user feedback. [9] The system (100) of claim 1, wherein the user interface module (110) with the display (112) is further configured to display detected skin conditions with spatial markers, visualize defect mapping results, display comparative model recognition results, and enable data entry of digital facial images and model selection preferences.

Images