Intelligent toilet diagnostic system

Abstract

An intelligent diagnostic toilet system for automated stool analysis, consisting of: a diagnostic chamber integrated into the drainage path of a toilet system, wherein the chamber is equipped with a hydrophobic and oleophobic optical window made of quartz or coated acrylic material, configured to allow multispectral light transmission while preventing contamination; a multispectral optical sensor module positioned next to the diagnostic chamber and configured to emit and detect reflected or transmitted light in at least the visible, near-infrared (NIR) and ultraviolet A (UV-A) ranges to quantify stool components including water content, lipid content, hemoglobin and bile pigments; a fluidically coupled unit with the diagnostic chamber for the detection of volatile organic compounds (VOCs), comprising a variety of solid-state gas sensors selected from the group consisting of metal oxide semiconductor, conductive polymer and ion mobility sensors, and configured to measure gases containing ammonia, hydrogen sulfide, methane and volatile fatty acids; a chemical test system comprising an interchangeable cartridge assembly containing a plurality of reagent-impregnated filter paper strips, wherein the cartridge is connected to a swab actuator configured to dispense a micro-amount of stool onto one or more reagent strips to trigger colorimetric chemical reactions; an imaging unit that is optically aligned with the aforementioned chemical test system to acquire multispectral colorimetric data of the reacted reagent strips and thus determine the presence of specific biochemical markers; a self-cleaning subsystem consisting of an array of micro-nozzle spray nozzles directed at the optical window and chamber surfaces for applying cleaning fluid, and a UV-C sterilization module located inside or near the chamber to decontaminate the surfaces after each analysis cycle; and an embedded processing unit operationally connected to the multispectral optical sensor module, the VOC detection unit and the imaging unit, wherein the processing unit performs a data fusion technique configured to correlate spectral, gaseous and colorimetric data to determine stool properties, generate diagnostic indices and output user-specific health assessments via an encrypted communication interface.

Description

Application area of ​​the invention:

[0001] The present invention relates to diagnostic health monitoring systems and, in particular, a smart toilet system that integrates non-ionizing multispectral optical sensors, arrays for the detection of volatile compounds, and filter paper-based chemical reaction modules for automated stool analysis. The invention further relates to a diagnostic device for point-of-care sample analysis, data acquisition, and AI-supported health assessment, enabling both clinical and traditional colorimetric analyses in the home or hospital setting. Background of the invention:

[0002] Current stool diagnostic systems are primarily based on manual sample collection and laboratory analysis, such as the guaiac test for occult blood. These methods are cumbersome, time-consuming, and pose hygiene risks. In parallel, various smart toilet systems have been developed that use image sensors and pH probes for the non-invasive analysis of stool characteristics such as color and consistency. However, these approaches lack biochemical validation and cannot detect minor abnormalities such as hidden blood, fat, or sugar.

[0003] A significant technological gap exists between automated multispectral scanners and conventional, chemical reaction-based tests. While manual tests are sensitive, they require manual execution by the user; sensor-based methods, although automated, can produce false positives without chemical confirmation. Therefore, there is a need for a hybrid system that combines optical spectroscopy, gas sensing, and chemical testing using filter paper on-site within a self-cleaning toilet infrastructure. The present invention addresses this need by integrating spectroscopic sensors with chemical colorimetry and automated AI interpretation to enable precise, hygienic, and comprehensive stool diagnostics.

[0004] Stool analysis remains one of the most informative and non-invasive diagnostic methods for assessing the health of the gastrointestinal tract, liver, pancreas, and metabolism. Stool contains a variety of biological and chemical markers, including bile pigments, undigested fats, proteins, residual blood, gut flora, and metabolic products. These parameters collectively reflect digestive efficiency, microbiome balance, and early signs of diseases such as malabsorption syndromes, ulcerative colitis, gastrointestinal bleeding, pancreatic insufficiency, and parasitic infections. However, conventional laboratory-based stool tests require manual sample collection and processing by trained personnel, which is inherently unhygienic and hinders user cooperation.Despite the clinical relevance of stool diagnostics, stool analysis is rarely included in routine examinations due to inconvenience, logistical effort, and the lack of simple automated devices that allow for hygienic and discreet multiparameter analysis.

[0005] Stool analysis is traditionally based on chemical, colorimetric tests performed manually using reagent-soaked filter paper or in solution. Common examples include the guaiac-based test for occult blood in stool, in which guaiac resin reacts with hemoglobin, producing a blue stain if hidden blood is present; Sudan III or Sudan IV staining to detect fatty acids and triglycerides, which indicate steatorrhea; and the Benedict reagent test to detect reducing sugars, which suggest carbohydrate malabsorption or intestinal infections. pH indicator strips are also used to determine acidity or alkalinity, thus revealing fermentation processes or bile acid deficiency. Although these methods are inexpensive and simple, they inevitably depend on correct sample collection, manual handling of the reagents, and subjective color interpretation by the individual.Furthermore, these tests cannot be performed automatically or in real time during defecation. They require stool to be collected in a container, sampled, and transferred to test paper under laboratory conditions. This not only poses a risk of contamination but also discourages people from frequent checkups, making such tests unsuitable for continuous or preventative healthcare.

[0006] In recent years, various concepts for so-called "smart toilets" have emerged in academic and industrial research, aiming to automate stool analysis using sensor-based methods. Some of these designs rely on image analysis with RGB cameras mounted inside the toilet bowl to analyze color, texture, and shape. These systems typically use computer vision and artificial intelligence to determine stool consistency based on the Bristol Stool Shape Scale, detect color abnormalities such as black or tarry stools that may indicate bleeding, and occasionally detect larger foreign bodies. While these systems represent progress toward automation, they rely solely on optical surface data, which is highly susceptible to lighting conditions, water turbidity, and surface reflections.The use of conventional cameras without controlled illumination or spectral filtering leads to inconsistent measurements. Furthermore, due to the lack of biochemical or molecular detection mechanisms, these systems are unable to identify occult blood, fat content, or sugar residues that are invisible to standard optical sensors.

[0007] Other approaches utilize near-infrared (NIR) or multispectral imaging techniques, inspired by systems used for materials analysis in agriculture and industry. These methods exploit the wavelength-dependent absorption and reflection properties of organic compounds to infer the presence of lipids, water, hemoglobin, and bile pigments. However, the practical application of NIR in toilets faces several technical challenges. The primary issue is optical fouling—the buildup of residue, moisture, or biofilm on the optical windows—which significantly distorts reflection measurements. High humidity and particles in toilets compromise the accuracy and lifespan of the sensors. Furthermore, the cost and calibration effort required for multispectral cameras or photodiodes add to the complexity.Most of the proposed systems remained in the laboratory stage due to these maintenance and reliability problems and could not be used in practice.

[0008] Attempts to measure stool odor or volatile organic compounds (VOCs) using electronic noses have also been described. These systems employ arrays of metal oxide or polymer composite gas sensors that detect ammonia, hydrogen sulfide, methane, and short-chain fatty acids released during stool decomposition. The concentration patterns of these gases correlate with bacterial metabolism and the health of the gut microbiome. While VOC analysis provides valuable biochemical insights, these systems alone cannot differentiate specific diseases because gas emissions are influenced by environmental factors such as humidity, ventilation, and waterproofing. Furthermore, gas sensors often exhibit drift over time, necessitating regular calibration.Therefore, stand-alone VOC measurement systems cannot provide reliable diagnostic conclusions without supplementary data from optical or chemical tests.

[0009] Some prototype smart toilets are experimenting with sampling mechanisms for more targeted biochemical analyses. For example, miniaturized robotic arms or rotating swabs have been developed to collect a stool sample and transfer it into microfluidic chambers or onto reagent pads. While these concepts are promising, they present mechanical complexity and hygiene concerns. Contact between moving parts and biological waste increases the risk of contamination and mechanical failure. Maintenance for such systems is high, and user trust and acceptance are low due to privacy and hygiene concerns. Furthermore, sample transfer introduces a time delay and additional potential sources of error compared to direct analysis within the waste stream.

[0010] Another significant limitation of existing smart toilet systems lies in their inability to correlate optical and chemical data. While a purely spectroscopic system can detect color and wavelength variations, it cannot definitively determine whether the red coloration originates from hemoglobin, food pigments, or bile. Chemical tests, such as the guaiac or Sudan test, provide definitive results but require refilling reagents and direct contact with the sample. No currently commercially available or research-based system fully integrates these two modalities into a unified system that enables real-time optical pre-screening followed by automated, reagent-based confirmation. Furthermore, the lack of data fusion prevents cross-validation, which is crucial for minimizing false positives and false negatives in diagnostic interpretation.

[0011] Data privacy and security concerns further restrict the widespread use of smart toilets. Systems that utilize ionizing radiation, such as dual-energy X-ray absorptiometry systems, are effective at material differentiation, but are unsuitable for use in homes or clinics due to the radiation exposure. Even non-ionizing optical systems face data privacy challenges when image data is stored or transmitted to external servers. Users are often hesitant to allow cloud-based image analysis of excrement because this data is highly personal. Therefore, an ideal diagnostic toilet must not only be non-invasive and non-ionizing, but also enable data protection-compliant, in-device data processing with encrypted result transmission.

[0012] From a maintenance perspective, optical clarity and hygiene continue to pose significant technical challenges. Biological waste environments are prone to contamination, microbial growth, and odor formation. Any optical or chemical sensor element exposed to such environments must have self-cleaning and sterilization mechanisms. Conventional systems often neglect these aspects, leading to reduced performance over time. Some designs suggest manual cleaning or the application of disinfectant sprays by the user, but these approaches defeat the purpose of automation. For long-term usability, a smart diagnostic toilet must incorporate self-cleaning nozzles, hydrophobic or oleophobic coatings, and UV-based sterilization to ensure durability and minimal user intervention.

[0013] In the field of data analysis, research on smart toilets largely focuses on narrowly defined aspects—either physical stool morphology or limited colorimetric analysis—without integrating multidimensional features such as spectral absorbance, VOC composition, and chemical reaction colorimetry. Consequently, diagnostic procedures remain superficial, relying on predefined thresholds or simple image processing rather than multimodal AI models trained on diverse features. Furthermore, few systems attempt to establish personalized baseline values, learning user-specific stool patterns over time to detect subtle variations that might indicate early disease onset.Without such adaptive modeling, generalized thresholds may incorrectly classify dietary changes or temporary digestive alterations as pathological, which impairs user reliability and trust.

[0014] Another drawback of current technology lies in the discrepancy between biomedical and traditional diagnostic paradigms. While modern stool analysis focuses on biochemical parameters, traditional systems like Ayurveda emphasize qualitative characteristics such as color (Varna), odor (Gandha), consistency (Sārata), and the floating or sinking behavior (Avasthā) of the stool as indicators of the internal balance or imbalance of the body humors (Doshas). No existing smart toilet device attempts to link modern sensor data with such holistic frameworks. This misses the opportunity for integrative diagnostics, which could broaden cultural acceptance and applicability in regions where traditional medicine and modern healthcare coexist.

[0015] Existing stool analysis solutions—whether manual laboratory methods or sensor-based smart toilets—are reaching their limits, as they must balance accuracy, automation, hygiene, and data privacy. While manual chemical tests offer accuracy, they are neither automated nor convenient. Optical and VOC-based sensors enable automation but lack confirmatory biochemical specificity. Mechanical sampling devices are complex and pose contamination risks. Furthermore, all known systems still suffer from issues such as optical contamination, calibration drift, data privacy concerns, and a lack of cross-validation.Therefore, there is an urgent need for a unified system that combines the precision of reagent-based chemical analyses with the automation and non-invasiveness of multispectral and VOC sensors, integrated into a self-cleaning and privacy-friendly toilet design that enables real-time, multimodal stool analysis. Such a system would bridge the gap between traditional and modern diagnostic methods while overcoming the operational and hygiene limitations that have thus far prevented the widespread use of stool analysis for home health monitoring. Summary of the invention:

[0016] The invention relates to an automated diagnostic device for toilets that analyzes stool composition in real time using optical, chemical, and gas-sensory subsystems. The system is housed in a sealed, hydrophobic discharge chamber behind the toilet bowl, thus ensuring privacy and preventing contamination.

[0017] In one embodiment, the diagnostic system is configured to implement an Ayurvedic diagnostic procedure in which stool analysis, according to traditional Ayurvedic clinical principles, is performed as the primary diagnostic test. In this configuration, the multispectral optical sensor module, the volatile compound detection unit, and the reagent-based chemical test system function as objective quantification instruments for measuring classic stool parameters described in Ayurvedic literature. The system thus standardizes traditional qualitative examination methods into measurable diagnostic indices while preserving the fundamental structure of Ayurvedic stool assessment.

[0018] The system comprises a multispectral imaging unit with RGB, near-infrared (NIR), and fluorescence excitation sensors arranged around an optical window made of quartz or acrylic. These sensors capture the reflected and emitted light from the stool sample and detect water, fat, bile pigments, hemoglobin, and fiber remnants.

[0019] An electronic nasal module, consisting of an array of sensors for volatile organic compounds (VOCs), measures gases such as ammonia, hydrogen sulfide and methane to draw conclusions about metabolic and infection-related conditions.

[0020] A test cassette housed in the same optical housing contains disposable filter paper strips impregnated with reagents such as guaiac resin for the detection of occult blood, Sudan III dye for fat detection, Benedict's reagent for the detection of reducing sugars, ninhydrin for the detection of mucus or proteins, and universal pH indicators. The cassette is actuated by a micro-scraper arm that automatically applies a stool sample to the strip and positions it under an optical reader. The system's RGB and hyperspectral cameras then capture the resulting color changes, while the integrated processor classifies the result using AI-based pattern recognition.

[0021] The combined results from the spectroscopic, VOC, and colorimetric subsystems are merged into a unified health report using a data fusion technique that cross-checks anomalies and trends over time. The evaluation system can output parameters such as stool consistency, fat content, presence of occult blood, bile pigment changes, pH value, and trends in microbial activity. The device is capable of mapping these parameters to both modern biomedical and traditional Ayurvedic diagnostic systems to enable a comprehensive interpretation.

[0022] The present invention aims to provide an automated, intelligent toilet diagnostic system that integrates multispectral optical sensors, volatile compound detection, and filter paper analysis using chemical reagents into a single, self-contained unit, enabling real-time stool analysis. The invention aims to eliminate reliance on manual sample collection and laboratory testing by providing contactless and hygienic diagnostics directly in the toilet drain. It is intended to achieve high diagnostic accuracy comparable to clinical biochemical tests while protecting user privacy and ensuring automation and ease of use suitable for home or hospital settings.

[0023] Another important objective of the invention is to overcome the limitations of existing smart toilets, which rely solely on visual or spectral imaging without biochemical confirmation. The invention achieves this through the use of filter paper cartridges impregnated with reagents that automatically react with stool components such as hemoglobin, bile pigments, lipids, proteins, or sugars. These reactions produce specific color changes, which are then optically analyzed by the system's image sensors, thus providing chemical confirmation in addition to spectroscopic and gaseous indicators. By integrating physical, optical, and chemical methods, the system can validate data and significantly reduce diagnostic errors due to optical noise or ambiguous coloration.

[0024] The invention also aims to provide a self-cleaning and dirt-repellent diagnostic chamber that ensures the long-term reliability of measurements in humid and biologically active environments. The invention provides for a hydrophobic or oleophobic coating of the optical window surface and the integration of automated micro-nozzle spray systems for rinsing away residual residues, followed by UV-C sterilization to eliminate microbial contamination. This design maintains optical clarity, prevents odor formation, and ensures hygienic operation without user intervention.

[0025] A further objective of the invention is the development of a device that operates exclusively on non-ionizing principles, thus ensuring absolute safety and suitability for permanent home use. The multispectral imaging system utilizes LED-based illumination in the visible, near-infrared, and ultraviolet A ranges, thereby avoiding the radiation hazards associated with high-energy imaging techniques. The VOC sensors used are energy-efficient semiconductor devices, and all subsystems are electrically insulated and housed in IP67-certified enclosures to protect them from moisture and electrical hazards. The system thus provides a safe and robust diagnostic platform suitable for use in both residential and clinical settings.

[0026] A further objective of the invention is to provide an intelligent framework for data fusion and interpretation that synthesizes input data from multispectral, chemical, and gas sensor modules to create a unified diagnostic profile. The integrated AI model correlates reflectance spectra, color shifts of reagents, and concentration patterns of volatile compounds to identify stool abnormalities such as occult blood, fat malabsorption, pH imbalance, or parasitic infestation. Through continuous learning and user-specific baseline adjustment, the system can adapt to individual physiological variations and detect deviations that may indicate the onset of disease or nutritional imbalance.

[0027] A further objective of the invention is to provide an environmentally friendly and cost-effective diagnostic solution that minimizes reagent consumption and waste. The use of disposable microtest strips in a replaceable cartridge enables the targeted use of chemical reagents only when required by the AI-based method based on optical pre-screening results. This adaptive testing strategy reduces reagent consumption and ensures their storage and handling in a controlled, contamination-free environment. Furthermore, the flushable or self-destructing strip design allows for hygienic disposal without manual handling or exposure.

[0028] The invention aims to address the challenges of data protection and data security in intelligent diagnostic systems. To this end, the system processes all raw data locally in an embedded computing unit. Only encrypted summary results and trend indices are transmitted to external devices such as smartphones or hospital servers via secure communication protocols. No visual or personal data is stored or transmitted, thus ensuring complete user anonymity and maintaining trust in the system.

[0029] A further aim of the invention is to establish interoperability between modern biomedical diagnostics and traditional paradigms of health assessment. By mapping sensor outputs such as color, odor (VOC patterns), texture, and density to classic Ayurvedic parameters such as Varna, Gandha, and Sarata, the system offers a dual interpretive framework that enriches both clinical and traditional health analysis. This feature improves the cultural adaptability of the device and opens up new avenues for integrative health monitoring, particularly in regions where traditional medicine is prevalent.

[0030] Another objective of the invention is to provide a modular and maintenance-friendly architecture for the diagnostic toilet, enabling regular maintenance or component replacement without requiring disassembly of the entire structure. The device design includes removable optical modules, replaceable reagent cartridges, and pluggable VOC sensors, ensuring durability and easy upgradeability as sensor technology advances. The modular approach also allows for scalability—from simple home models with limited sensor capabilities to advanced clinical models with expanded diagnostic modules and AI analytics.

[0031] Ultimately, the invention aims to create a comprehensive tool for preventive healthcare that enables users to continuously monitor their gastrointestinal tract and metabolism, thus allowing for the early detection of pathological changes long before the onset of clinical symptoms. Through real-time feedback on nutrition, hydration, digestion, and intestinal health, the system promotes proactive health management. The integration of automated testing, chemical confirmation, and adaptive analytics positions the invention as a groundbreaking advancement in personal health technology, bridging the gap between laboratory diagnostics and everyday household hygiene devices. BRIEF DESCRIPTION OF THE IMAGE

[0032] These and other features, aspects and advantages of the present invention will be better understood if the following detailed description is read with reference to the accompanying drawing, in which the same symbols represent the same parts: Fig. Figure 1 shows a block diagram of an integrated intelligent toilet diagnostic system with multimodal spectroscopic, volatile compound and filter paper-based chemical test analysis.

[0033] Furthermore, those skilled in the art will recognize that the elements in the drawing are simplified and not necessarily drawn to scale. For example, the flowcharts illustrate the process by highlighting the main steps to facilitate understanding of the present disclosure. With regard to the construction of the device, one or more components may be represented in the drawing by conventional symbols. The drawing may show only those specific details relevant to understanding the embodiments of the present disclosure, so as not to clutter the drawing with details that are already apparent to those skilled in the art from the description contained herein. Detailed description of the invention

[0034] To facilitate understanding of the principles of the invention, reference is made below to the embodiment shown in the drawing, which is described using specific terms. It is understood, however, that this does not limit the scope of protection of the invention. Rather, modifications and further developments of the depicted system, as well as further applications of the inventive principles shown therein, are conceivable, insofar as they would normally occur to a person skilled in the art in the field of the invention.

[0035] It will be clear to those skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not to be understood as a limitation thereof.

[0036] References to “an aspect”, “another aspect”, or similar phrases in this description mean that a particular feature, structure, or property described in connection with the embodiment is included in at least one embodiment of the present disclosure. Therefore, phrases such as “in one embodiment”, “in another embodiment”, and similar expressions in this description may, but do not necessarily, all refer to the same embodiment.

[0037] The terms "includes," "comprehensive," or similar expressions denote non-exclusive inclusion. Thus, a procedure or method containing a list of steps does not only include those steps but may also include further steps not explicitly listed or inherent in the procedure or method. Likewise, the statement "includes..." for one or more devices, subsystems, elements, structures, or components, without further limitations, does not preclude the existence of other devices, subsystems, elements, structures, or components.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meanings generally known to those skilled in the art in the field to which this invention belongs. The systems, methods, and examples described herein serve only for illustration and are not to be understood as limiting.

[0039] Embodiments of the present disclosure are described in detail below with reference to the attached drawing.

[0040] Fig.Figure 1 shows a block diagram of an integrated intelligent toilet diagnostic system with multimodal spectroscopic, volatile, and filter paper-based chemical test analysis. The system 100 comprises: a diagnostic chamber (102) integrated into the drain of a toilet structure, featuring a hydrophobic and oleophobic optical window made of quartz or coated acrylic that transmits multispectral light while preventing fouling; a multispectral optical sensor module (104) located adjacent to the diagnostic chamber, configured to emit and detect reflected or transmitted light in at least the visible, near-infrared (NIR), and ultraviolet A (UV-A) ranges to quantify stool constituents such as water content, lipid content, hemoglobin, and bile pigments;a fluidically coupled unit (106) to the diagnostic chamber for the detection of volatile organic compounds (VOCs), comprising several solid-state gas sensors selected from the group consisting of metal oxide semiconductor, conductive polymer, and ion mobility sensors configured to measure gases such as ammonia, hydrogen sulfide, methane, and volatile fatty acids; a chemical test system (108) with an exchangeable cartridge assembly containing several reagent-impregnated filter paper strips, the cartridge being connected to a swab actuator (108a) that applies a microquantity of stool to one or more reagent strips to trigger colorimetric chemical reactions; an imaging unit (110) optically aligned with the chemical test system to acquire multispectral colorimetric data of the reacted reagent strips and thus determine the presence of specific biochemical markers;a self-cleaning subsystem (112) comprising an array of micronozzle spray jets directed at the optical window and chamber surfaces to dispense cleaning fluid, and a UV-C sterilization module located within or near the chamber to decontaminate surfaces after each analysis cycle; and an embedded processing unit (114) operationally linked to the multispectral optical sensor module, the VOC detection unit, and the imaging unit, wherein the processing unit performs a data fusion technique configured to correlate spectral, gaseous, and colorimetric data to determine stool properties, generate diagnostic indices, and output user-specific health assessments via an encrypted communication interface.

[0041] In one embodiment, the multispectral optical sensor module (104) comprises an array of controlled illumination sources, including visible-range white LEDs, narrowband NIR LEDs in the wavelength range of 700–1050 nm, and UV-A excitation sources in the range of 365–400 nm, coupled with a plurality of photodiode detectors equipped with wavelength-selective optical filters to detect reflected intensities corresponding to the absorption peaks of hemoglobin, the spectral bands of lipids, and the signatures of bile pigments. The module operates in sync with the rinsing cycle to minimize optical interference caused by turbulence or water flow.

[0042] In one embodiment, the diagnostic chamber (102) further comprises an internal geometry designed to maintain laminar flow of the waste material. It is characterized by a constricted viewing section with a length of 50-120 mm and a cross-sectional area ratio between the inlet and the viewing zone of at least 2:1, thereby ensuring stable optical measurement. The inner walls are coated with a fluoropolymer-based non-stick layer to minimize the adhesion of organic residues.

[0043] In one embodiment, the VOC detection unit (106) is enclosed in a sealed microchannel, which is connected to the diagnostic chamber via a unilaterally gas-permeable membrane. The membrane consists of a polytetrafluoroethylene (PTFE) or silicone layer, which allows selective diffusion of volatile gases while blocking aerosols or liquids. This maintains the sensor's longevity and measurement accuracy in humid environments.

[0044] In one embodiment, the chemical test subsystem (108) is configured such that each reagent-impregnated filter paper strip is stored in an insulated compartment of the cartridge to prevent cross-contamination. Each compartment is preloaded with reagents including guaiac resin and hydrogen peroxide for the detection of occult blood, Sudan III or Sudan IV for lipid analysis, Benedict's solution for the identification of reducing sugars, ninhydrin for the detection of mucus or proteins, and universal indicators for pH determination. The cartridge features an electromechanically actuated indexing wheel for the sequential feeding of the test strips to the swab actuator.

[0045] In one embodiment, the swab actuator (108b) comprises a microservo-driven arm terminating in a sterilizable elastomeric pad configured to contact the stool stream for less than two seconds to collect a controlled microsample of approximately 0.05 to 0.1 grams. The actuator is enclosed in a retractable housing equipped with a hydrophobic sealing membrane that prevents the ingress of liquid or aerosol when inactive.

[0046] In one embodiment, the imaging unit (110) comprises a high dynamic range CMOS camera with a spectral sensitivity of 400-1000 nm, which is integrated with a tunable optical filter or a diffractive spectral separator to enable the acquisition of multiband color images of reacted reagent strips, wherein the imaging unit is calibrated using internal reference fields to compensate for fluctuations in LED intensity and the aging of the reagents.

[0047] In one embodiment, the self-cleaning system (112) operates in a multi-stage cycle comprising: (a) activation of precision micro-nozzle spray jets that deliver a cleaning fluid consisting of sterile water and low-concentration disinfectant at a pressure between 1 and 2 bar; (b) drainage of the residual fluid via a gravity-assisted drain channel integrated into the chamber floor; (c) activation of a blower or heating element for drying; and (d) irradiation with UV-C radiation at wavelengths between 250 and 280 nm for a duration of at least 30 seconds, thereby ensuring complete sterilization of the optical surfaces and the removal of microbial residues before the next use.

[0048] In one embodiment, the embedded processing unit (114) comprises a microcontroller coupled with a dedicated AI coprocessor or neural accelerator that executes trained machine learning models configured to perform multispectral feature extraction, spectral deconvolution, and decision-level data fusion across optical, chemical, and gaseous modalities, wherein the AI ​​model further performs adaptive baseline establishment by storing temporal stool profiles of an individual user and adjusting diagnostic thresholds based on longitudinal trends.

[0049] In one embodiment, the data fusion technique implemented by the embedded processing unit performs a joint probabilistic correlation between optical reflection spectra, colorimetric reactions of reagent strips, and VOC concentration vectors, wherein a confidence-weighted inference model assigns variable importance factors to each modality based on the environmental conditions, so that if the quality of the optical data is impaired due to turbidity, a higher diagnostic weight is assigned to the reagent or VOC results to ensure the reliability of the output health indices.

[0050] In one embodiment of the present invention, the diagnostic system is fundamentally configured to perform the classical Ayurvedic stool examination as the primary diagnostic test. In this configuration, the system implements an Ayurvedic diagnostic concept based on traditional clinical teachings described in authoritative Ayurvedic literature, including classical compendiums of internal medicine and surgery.

[0051] In this embodiment, stool analysis serves as the primary diagnostic method for assessing digestive function, metabolic efficiency, and systemic physiological balance. The system is configured to evaluate parameters corresponding to classic stool examination characteristics such as color, odor, consistency, density, buoyancy, presence of mucus, and undigested residue.

[0052] Unlike systems that employ traditional analytical methods as a secondary interpretation stage, the present invention establishes the Ayurvedic diagnostic framework as the primary examination method. The multispectral optical sensor module, the volatile organic compound detection unit, and the chemical reagent system serve as objective quantification mechanisms that convert traditionally qualitative stool characteristics into standardized, measurable indices.

[0053] The integrated processing unit executes a structured evaluation protocol in which classic stool analysis parameters are converted into quantitative feature sets using calibrated sensor measurements. These quantified features are then correlated within the framework of Ayurvedic diagnostics to generate standardized indices representing digestive power, metabolic balance, and the presence of pathological residues.

[0054] By establishing stool analysis as the primary diagnostic procedure, the system formalizes and digitizes traditional clinical assessment into a reproducible, sensor-controlled format while ensuring compliance with modern biomedical measurement standards.

[0055] In one embodiment of the present invention, the system is configured to evaluate classical stool examination parameters as primary diagnostic test variables within an Ayurvedic diagnostic model. These parameters are treated as measurable clinical criteria and not as interpretive results. The integrated processing unit, in conjunction with the multispectral optical sensor module, the volatile compound detection unit, the chemical test system, and the flow measurement architecture, performs an objective quantification of the following parameters: stool color by means of calibrated spectral reflectance analysis; odor properties by means of volatile organic compound concentration vectors; consistency and compactness by means of texture profiling and structural cohesion indices; and the presence of mucus or surface stickiness by means of protein-reactive chemical reactions and surface reflectance measurements.Oiliness is assessed using lipid-sensitive spectral bands and fat-reactive reagent strips; dryness is assessed using moisture absorption ratios and dehydration signatures; relative heaviness or lightness is assessed using density estimation and flotation behavior in the diagnostic chamber; presence or absence of metabolic waste products is assessed using correlated chemical and gaseous biomarker patterns; excretion ease and frequency are assessed using event timing and flow monitoring; and structural form and fragmentation patterns are assessed using image-based morphology analysis, with each parameter objectively quantified and processed using calibrated sensor data to generate standardized diagnostic indices within the Ayurvedic diagnostic framework.

[0056] To digitize, standardize, and objectively quantify the stool examination parameters defined within Ayurvedic diagnostics, the system uses a variety of coordinated sensor and chemical analysis subsystems, each configured to correspond to specific classical diagnostic variables through measurable physical and biochemical indicators. Stool color is objectively quantified using multispectral optical sensors that include visible RGB imaging, near-infrared reflection, and UV-induced fluorescence. Odor properties are quantified using a volatile organic compound (VOC) detection array that includes ammonia, hydrogen sulfide,Methane and short-chain fatty acids are measured. The presence of mucus or abnormal surface stickiness is determined by protein- or mucus-reactive chemical reagents in combination with surface reflectance analysis. Consistency and compactness are assessed by optical texture profiling, evaluation of structural cohesion, and monitoring of the flow pattern in the diagnostic chamber. Oil content is quantified using lipid-binding colorimetric reagents and near-infrared lipid absorption bands. Dryness is determined by near-infrared water absorption characteristics and moisture ratio calculation; metabolically undigested stool states are identified by correlated acidic pH values, elevated concentrations of fermentation gases, and abnormal volatile compound profiles, while properly processed stool states are indicated by a near-neutral pH.characterized by a low concentration of volatile compounds and stable spectral features; heavier stool properties are associated with a higher lipid content, increased density indices, and delayed structural dispersion; and lighter stool properties are associated with a lower residue content, higher water content, and rapid dispersion dynamics, with all these parameters being derived from calibrated sensor outputs and integrated by the embedded processing unit to generate standardized diagnostic indices within the Ayurvedic diagnostic framework.

[0057] In one embodiment of the present invention, the diagnostic sequence is structured such that stool analysis according to the Ayurvedic diagnostic model is performed as the primary diagnostic test. The multispectral optical sensor module, the volatile organic compound detection unit, and the chemical reagent-based test system are activated as objective quantification instruments to measure the classical stool analysis parameters defined in this model. The integrated processing unit receives calibrated, sensor-based data and processes it to generate quantified representations of classical diagnostic features, rather than creating independent, modern diagnostic labels detached from the underlying model.The AI-based inference engine serves to standardize the stool analysis process, minimizes the observer variability typical of manual assessment, and enables the longitudinal tracking of quantified stool parameters over time for trend analysis. By structuring the system in this way, the invention achieves the digitization and objective standardization of classical stool analysis, integrates traditional diagnostics with modern sensor technologies without reducing the traditional framework to a secondary interpretive output, enables reproducible and preventive health-oriented assessments, and establishes a unified diagnostic platform in which the traditional framework forms the basic diagnostic logic, while modern sensor subsystems ensure measurable validation and technical precision.

[0058] The intelligent diagnostic toilet system described here utilizes a complex interplay of optical, chemical, mechanical, and computer-aided subsystems to enable fully automated, real-time stool analysis. At its core is a hybrid diagnostic framework that integrates multispectral imaging, volatile organic compound (VOC) detection, and colorimetric chemical tests. The results are fed into a unified technical evaluation engine implemented on an embedded processor. The architecture is designed for cross-domain data acquisition, data harmonization, cross-validation, and adaptive learning, while simultaneously ensuring hygienic operation and complete user privacy.

[0059] In one embodiment of the present invention, the imaging unit and the integrated processing unit are configured to perform an automated classification of stool morphology based on the structural shape, segmentation features, and surface texture of the stool sample captured in the diagnostic chamber. The multispectral imaging unit acquires high-resolution images of the stool as it passes through the diagnostic chamber, and the integrated processing unit extracts morphological features such as contour length, segmentation discontinuities, surface smoothness, edge irregularities, and volumetric dispersion patterns.

[0060] Based on these extracted features, the system classifies stool consistency into several diagnostic morphological categories that correspond to clinically recognized stool shape classifications. These classification categories include: Type I morphology, characterized by several separate, hard, pellet-like segments, indicating severe constipation and reduced intestinal motility; Type II morphology, characterized by soft but clearly demarcated, rounded segments with defined edges, indicating low fiber intake and incomplete stool cohesion; Type III morphology, characterized by an elongated, sausage-shaped structure with visible segmentation or clumping on the surface, corresponding to mild obstruction; Type IV morphology, characterized by a continuous cylindrical structure with a smooth and soft surface and uniform diameter, representing a physiologically normal stool consistency and optimal gastrointestinal passage; Morphology of type V, characterized by a cylindrical stool structure with surface cracks while maintaining overall shape integrity, also corresponds to a normal but slightly dehydrated stool condition; Morphology of type VI, characterized by loosely aggregated fragments with irregular edges and partial structural breakdown, indicating mild diarrheal disease and increased fluid content in the intestine; Morphology of type VII, characterized by predominantly liquid stools with minimal or no solid structure, indicating severe diarrheal disease and rapid gastrointestinal passage.

[0061] The integrated processing unit performs quantitative morphology classification using computer vision techniques, including contour recognition, curvature analysis, surface texture analysis, and fragmentation indexing. Morphological parameters such as segment count, elongation ratio, surface smoothness index, and structural integrity coefficient are calculated and assigned to stool morphology categories. The classification results are integrated with the multispectral optical sensor module, the volatile organic compound (VOC) detection unit, and the reagent-based chemical assay system to enable a comprehensive assessment of gastrointestinal health. Morphology classification also contributes to diagnostic indicators related to intestinal transit time, hydration status, dietary fiber intake, and potential gastrointestinal diseases.

[0062] The diagnostic process begins as soon as the user flushes the toilet. The control module initiates a synchronized sampling procedure, during which the stool flows through a specially designed diagnostic chamber located behind the toilet bowl. This chamber is equipped with an optically transparent window made of quartz or coated acrylic glass, featuring hydrophobic and oleophobic coatings to prevent deposits and optical distortion. The chamber's internal geometry is designed to maintain laminar flow and reduce turbulence during the brief analysis time, ensuring stable reflectance measurements. As the stool sample passes through the chamber, an array of multispectral light sources emits controlled pulses of light in the visible, near-infrared (NIR), and ultraviolet (UV-A) spectral ranges.

[0063] The multispectral imaging system consists of multiple photodiodes or a CMOS image sensor array combined with wavelength-selective optical filters. These isolate narrow spectral bands corresponding to the absorption characteristics of hemoglobin, lipids, bile pigments, and water. Specifically, the system monitors spectral intensities in the 540–580 nm range for oxyhemoglobin absorption, 920–970 nm for lipid and moisture content, and 400–450 nm for bile and bilirubin fluorescence under UV-A excitation. Each illumination-emission cycle is precisely time-synchronized with the partial flushing phase to minimize the influence of water turbidity. The acquired optical signal is digitized and preprocessed by the integrated controller for normalization, background subtraction, and illumination correction.

[0064] In parallel with the optical measurement, a special VOC detection system analyzes the gas in the headspace of the diagnostic chamber through a one-way gas permeability membrane. This microchannel contains an array of MEMS-based metal oxide and conductive polymer gas sensors, each responding to different analytes such as ammonia, hydrogen sulfide, methane, and volatile short-chain fatty acids. The gas sensors generate analog voltage signals proportional to the concentration of the respective compound, which are digitized and transmitted to the processor. Because gas sensors are prone to drift, the system performs a dynamic baseline calibration during idle periods by comparing the measured values ​​with the ambient air and applying compensation factors stored in non-volatile memory.The VOC data provide important biochemical information that complements the optical measurements and allows differentiation between infections, fermentation states or microbiome imbalances that might not be detectable based on spectral data alone.

[0065] A key feature of this invention is the integration of a reagent-based chemical test system that provides biochemical confirmation. At the command of the control processor, a microservo-driven actuator extends a sterilizable, elastic swab into the stool stream for less than two seconds to collect a microsample of controlled volume (typically 50 to 100 milligrams). The actuator then retracts and dispenses the sample onto a selected, reagent-impregnated filter paper strip in a sealed cartridge. Each strip is preloaded with a specific reagent, such as guaiac for the detection of occult blood, Sudan III or IV for the detection of fat, Benedict's reagent for the detection of reducing sugars, ninhydrin for the detection of proteins and mucus, and universal indicators for pH determination.After a predefined reaction time, the reagent strip is positioned under the multispectral camera, which captures colorimetric images in the visible and near-infrared range. The captured image undergoes spectral decomposition, and the resulting chromatic data are quantified in terms of hue, saturation, intensity, and reflectance to determine the presence and extent of chemical reactions.

[0066] The fusion of multispectral, chemical, and gaseous data streams is performed by an embedded AI coprocessor. The data processing technique comprises four main stages: feature extraction, modality-specific preprocessing, probabilistic data fusion, and adaptive learning. In feature extraction, optical data is transformed into spectral fingerprints by calculating intensity ratios at diagnostic wavelengths. For example, the ratio of intensities at 570 nm to 610 nm corresponds to the hemoglobin concentration, while the reflectance at 940 nm relative to 850 nm indicates the lipid content. VOC signals are transformed into normalized concentration vectors using logarithmic transformation and temperature-humidity compensation models.Chemical stripe images are processed by a deep convolutional neural network (CNN) trained with labeled reaction results to quantify colorimetric intensities and predict a positive test result with statistical confidence.

[0067] Once all feature vectors are extracted, the data fusion module aligns them temporally and applies a probabilistic correlation model based on Bayesian inference. Each measurement modality is assigned a confidence weight that varies depending on environmental conditions and data quality metrics. For example, the confidence weight of the optical subsystem is reduced if it detects excessive haze or reflection anomalies, while reagent-based and VOC data are weighted more heavily. Conversely, the process dynamically shifts the diagnostic weight toward the spectral and gaseous subsystems when the reagent cartridge reaches the end of its lifespan or the color saturation falls below a threshold. This adaptive weighting ensures robust conclusions even under suboptimal measurement conditions.

[0068] The inference model operates as a hierarchical ensemble of classifiers, combining gradient-boosting decision trees and probabilistic neural networks. The primary classifier distinguishes between normal and abnormal stool states based on spectral and chemical features, while secondary classifiers identify specific anomaly categories such as steatorrhea, occult bleeding, acidic stool, or microbial fermentation. These models are pre-trained using large datasets of synthetic stool phantoms and clinically verified samples and continuously optimized through unsupervised, user-specific adjustments. The adaptive learning component employs a temporal smoothing mechanism to establish personalized baselines for each user by storing mean and variance profiles of spectral and VOC patterns across multiple measurements.Subsequent measurements are compared to these baselines, and only statistically significant deviations trigger warnings. This individualized calibration significantly improves diagnostic precision and reduces the rate of false positives due to dietary or lifestyle changes.

[0069] The entire diagnostic process is orchestrated by event-driven control technology, which coordinates sensor activation, actuator control, and cleaning operations. The system operates a closed-loop control system that monitors optical clarity, reagent response time, and VOC signal stability, adjusting illumination intensity, exposure time, and actuator pressure as needed. Upon completion of the diagnostic cycle, the chamber initiates a self-cleaning sequence using pressurized water jets to clean the optical window and interior walls. The liquid is drained via gravity channels, followed by warm air drying and UV-C sterilization for at least thirty seconds to eliminate microbial residues.

[0070] Data security is an integral part of the system architecture. The embedded processor performs all raw data analysis locally, ensuring that only encrypted summary indices are transmitted externally. The communication interface uses AES-256 encryption or equivalent standards to prevent unauthorized access. The diagnostic summaries are formatted as anonymized health indices, indicating parameters such as body fat percentage, blood fraction, and pH deviations without transmitting any visual or identifiable information. Results can be viewed via a local interface or a dedicated smartphone app. The user receives trend charts and alerts for deviations from baseline values.

[0071] The system also includes a hybrid interpretation layer that maps modern biochemical parameters to traditional diagnostic features derived from Ayurveda. The AI ​​model incorporates a rule-based logic module that correlates stool color, odor, and consistency with descriptions such as Varna (color), Gandha (odor), and Särata (consistency), thus enabling integrative health feedback that aligns with both modern medicine and traditional holistic approaches.

[0072] Through the combined application of advanced spectroscopy, chemical confirmation, gas analysis, and machine learning, the invention offers a comprehensive and self-contained diagnostic system. It not only detects clinically relevant stool changes with high specificity but also continuously adapts to the user's physiology and environmental factors. The system's detailed technical and structural synergy ensures consistent performance, minimal maintenance, and maximum diagnostic reliability. This transforms routine toilet use into a safe and hygienic form of health monitoring, enabling early disease detection and continuous assessment of well-being.

[0073] The diagnostic system is integrated into the drainage area of ​​a toilet and consists of a sealed diagnostic chamber with a transparent window made of hydrophobic, dirt-repellent quartz glass. The chamber is oriented in the direction of flow, so that the stool passes through it for approximately two to five seconds before the flush is finally triggered. Inside the chamber are multispectral light sources, including LED emitters for visible light, near-infrared, and UV-A, as well as corresponding photodiode detectors for measuring reflection.

[0074] A self-cleaning mechanism with micro-nozzles and UV-C sterilization lamps is arranged around the optical window. After each use, the chamber undergoes a rinsing cycle with sterilization solution or high-pressure water jets, followed by air drying to maintain optical clarity. The chamber surface is coated with a hydrophobic polymer film to prevent deposits. The VOC sensor unit is located in an adjacent microchannel that collects a controlled sample of the gases produced during or after defecation. This microchannel contains an array of MEMS-based metal oxide gas sensors that measure the concentrations of ammonia, hydrogen sulfide, and methane.

[0075] The chemical test module consists of a cartridge compartment containing several pre-loaded filter paper strips, each impregnated with a specific chemical reagent. On command, a servo-controlled arm removes a test strip and exposes it to a controlled amount of stool material collected via a swab mechanism. After exposure, the strip is inserted into the scanning area below the optical sensor array, where the color reaction is recorded by the multispectral imaging unit. The used strip is then disposed of in a sealed, flushable compartment.

[0076] The system's processing unit comprises a microcontroller and an integrated AI model that performs colorimetric analysis, spectral feature extraction, and decision fusion. The device utilizes a method analogous to dual-energy X-ray material discrimination, but employs multiple non-ionizing optical spectral bands to differentiate between water, lipid, and hemoglobin components. A calibration routine using synthetic stool phantoms of varying composition (water, fat, and bile pigments) ensures consistent performance across different users and installations.

[0077] The collected data is encrypted and processed locally to protect user privacy. Only summary health indicators are transmitted to a corresponding mobile app or clinical monitoring application. Optional integration with cloud-based health platforms enables long-term monitoring and telemedicine support.

[0078] After bowel movement, the system initiates a partial flush to stabilize the sample in the diagnostic chamber. The multispectral imaging module captures the initial reflectance data. Simultaneously, VOC sensors measure the gas composition. The micro-arm applies a stool sample to one or more chemical filter papers, triggering specific reactions. Once the reaction is complete, the RGB camera records the color changes. The AI ​​model compares the chemical results with the spectral and VOC data. If all three subsystems confirm an abnormal reading, such as occult blood or excess fat, the device sends a diagnostic alert to the connected user app.

[0079] The drawing and the preceding description illustrate embodiments. Those skilled in the art will recognize that one or more of the described elements can be combined to form a single functional element. Alternatively, certain elements can be divided into several functional elements. Elements of one embodiment can be added to another. For example, the process flows described here can be modified and are not limited to the manner described herein. Furthermore, the actions of a flowchart need not be performed in the sequence shown; nor do all actions necessarily need to be carried out. Actions that do not depend on other actions can be performed in parallel with the other actions. The scope of protection of the embodiments is in no way limited by these specific examples. Numerous variations, whether explicitly stated in the description or not, such as...Differences in structure, dimensions, and materials are possible. The scope of protection of the embodiments is at least as comprehensive as described by the following claims.

[0080] The advantages, other benefits, and problem solutions have been described above with reference to specific embodiments. However, the advantages, benefits, problem solutions, and any components that can effect or enhance an advantage, benefit, or solution are not to be construed as critical, necessary, or essential features or components of the claims. REFERENCES 100 An integrated intelligent toilet diagnostic system with multimodal spectroscopic analysis, volatile compound analysis and chemical test analysis on filter paper. 102 Diagnostic Chamber 104 Multispectral Optical Sensor Module 106 Detection Unit for Volatile Organic Compounds (VOCs) 108 Subsystem for Chemical Tests 108a Smear actuator 110 Imaging Unit 112 Self-cleaning system 114 Embedded processing unit

Claims

An intelligent diagnostic toilet system for automated stool analysis, consisting of: a diagnostic chamber integrated into the drainage path of a toilet system, wherein the chamber is equipped with a hydrophobic and oleophobic optical window made of quartz or coated acrylic material, configured to allow multispectral light transmission while preventing contamination; a multispectral optical sensor module positioned next to the diagnostic chamber and configured to emit and detect reflected or transmitted light in at least the visible, near-infrared (NIR) and ultraviolet A (UV-A) ranges to quantify stool components including water content, lipid content, hemoglobin and bile pigments;a fluidically coupled unit to the diagnostic chamber for the detection of volatile organic compounds (VOCs), comprising a variety of solid-state gas sensors selected from the group consisting of metal oxide semiconductor, conductive polymer, and ion mobility sensors, configured to measure gases containing ammonia, hydrogen sulfide, methane, and volatile fatty acids; a chemical test system comprising an exchangeable cartridge assembly containing a variety of reagent-impregnated filter paper strips, the cartridge being connected to a swab actuator configured to dispense a micro-quantity of stool onto one or more reagent strips to trigger colorimetric chemical reactions;an imaging unit optically aligned with the aforementioned chemical test system to acquire multispectral colorimetric data of the reacted reagent strips and thus determine the presence of specific biochemical markers; a self-cleaning subsystem consisting of an array of micronozzle spray nozzles directed at the optical window and chamber surfaces for the application of cleaning fluid, and a UV-C sterilization module located within or near the chamber to decontaminate the surfaces after each analysis cycle;and an embedded processing unit operationally connected to the multispectral optical sensor module, the VOC detection unit, and the imaging unit, wherein the processing unit performs a data fusion technique configured to correlate spectral, gaseous, and colorimetric data to determine stool characteristics, generate diagnostic indices, and output user-specific health assessments via an encrypted communication interface. System according to claim 1, wherein the multispectral optical sensor module comprises an arrangement of controlled illumination sources, including LEDs for white light in the visible range, narrowband NIR LEDs in the wavelength range of 700-1050 nm and UV-A excitation sources in the range of 365-400 nm, coupled with a plurality of photodiode detectors equipped with wavelength-selective optical filters to detect reflected intensities corresponding to the absorption peaks of hemoglobin, the spectral bands of lipids and the signatures of bile pigments, wherein the module operates in time synchronization with the rinsing cycle to minimize optical interference caused by turbulence or water flow. System according to claim 1, wherein the diagnostic chamber further comprises an internal geometry designed to maintain a laminar flow of the waste material, characterized by a narrowed viewing section with a length of 50-120 mm and a cross-sectional area ratio between inlet and viewing zone of at least 2:1, thereby ensuring stable optical measurement, and wherein the inner walls are coated with a fluoropolymer-based non-stick layer to minimize the adhesion of organic residues. System according to claim 1, wherein the VOC detection unit is enclosed in a sealed microchannel which is connected to the diagnostic chamber via a unilaterally gas-permeable membrane, wherein the membrane comprises a polytetrafluoroethylene (PTFE) or silicone layer which allows selective diffusion of volatile gases while blocking aerosols or liquids, thereby maintaining the longevity of the sensor and the measurement accuracy in a humid environment. System according to claim 1, wherein the chemical test subsystem is configured such that each reagent-impregnated filter paper strip is stored in an isolated compartment of the cartridge to avoid cross-contamination, wherein each compartment is preloaded with reagents comprising guaiac resin and hydrogen peroxide for the detection of occult blood, Sudan III or Sudan IV for lipid analysis, Benedict's solution for the identification of reducing sugars, ninhydrin for the detection of mucus or proteins, and universal indicator dyes for pH determination, and wherein the cartridge has an electromechanically actuated indexing wheel for the sequential feeding of test strips to the swab actuator. System according to claim 1, wherein the swab actuator comprises a microservo-driven arm terminating in a sterilizable elastomeric pad configured to contact the stool stream for less than two seconds to obtain a controlled microsample of about 0.05 to 0.1 grams, wherein the actuator is enclosed in a retractable housing equipped with a hydrophobic sealing membrane that prevents the ingress of liquid or aerosol in the inactive state. System according to claim 1, wherein the imaging unit comprises a high dynamic range CMOS camera with a spectral sensitivity of 400-1000 nm, which is integrated with a tunable optical filter or a diffractive spectral separator to enable the acquisition of multiband color images of reacted reagent strips, and wherein the imaging unit is calibrated using internal reference fields to compensate for fluctuations in LED intensity and the aging of the reagents. System according to claim 1, wherein the embedded processing unit comprises a microcontroller coupled to a dedicated AI coprocessor or neural accelerator executing trained machine learning models configured to perform multispectral feature extraction, spectral deconvolution, and decision-level data fusion across optical, chemical, and gaseous modalities, wherein the AI ​​model further performs adaptive baseline establishment by storing temporal stool profiles of a single user and adjusting diagnostic thresholds based on longitudinal trends. System according to claim 1, wherein the data fusion technique implemented by the embedded processing unit performs a common probabilistic correlation between optical reflectance spectra, colorimetric reactions of reagent strips and VOC concentration vectors, and wherein a confidence-weighted inference model assigns variable importance factors to each modality based on environmental conditions, such that if the quality of the optical data is impaired due to turbidity, a higher diagnostic weight is assigned to the reagent or VOC results to ensure the reliability of the health indices issued.

Images