Methods, systems and computer programs using machine learning to optimize prediction of an occurrence of a recurring medical symptom or body behavior

Abstract

A computer-implemented method for optimizing a prediction of an occurrence of a recurring medical symptom or body behavior of a human being using machine learning, the method comprising obtaining (110) input data comprising features related to the occurrence of the recurring medical symptom or body behavior, dividing (120) the features included in the input data into two or more groups of features, encoding (130), using a trained encoder machine learning model, the features of the two or more groups of features into two or more embeddings, with each embedding representing a group of features, inputting (150) the two or more embeddings into a prediction machine learning model being trained to predict the occurrence of a recurring medical symptom or body behavior based on the two or more embeddings, and providing (160) a prediction of the recurring medical symptom or body behavior based on an output of the prediction machine learning model.

Description

[0001] METHODS, SYSTEMS AND COMPUTER PROGRAMS USINGMACHINE LEARNING TO OPTIMIZE PREDICTION OF AN OCCURRENCE OF A RECURRING MEDICAL SYMPTOM OR BODY BEHAVIOR The present invention relates to a computer-implemented method for optimizing a prediction of an occurrence of a recurring medical symptom or body behavior of a human being using machine learning, to a computer-implemented method for training at least one machine learning model for use in predicting occurrence of a recurring medical symptom or body behavior, and to corresponding systems, computer programs and non-transitory, computer-readable media. Migraine is the most prevalent neurological disorder affecting 12% of the global population. It is influenced by multiple contributing factors, including dietary habits, daily routines, genetics, as well as physiological and biochemical factors. Migraine manifests as recurring, incapacitating headache episodes, occurring either sporadically or persistently, with or without accompanying sensory disturbances. Migraine ranks as the second most significant contributor to the global disability, primarily affecting individuals between the productive ages of 22 and 55, with a higher prevalence among women. This condition represents a substantial source of economic losses on a global scale. However, predicting migraine occurrences and identifying triggers for individual patients remain challenging tasks. Predicting migraine occurrences and identifying triggers remain challenging due to the wide variability in individual triggers, which can include stress, hormonal changes, certain foods, environmental factors, and sleep patterns (see Khan, Johra, et al. “Genetics, pathophysiology, diagnosis, treatment, management, and prevention of migraine”). The complexity is compounded by the cumulative and inconsistent nature of these triggers, individual physiological differences, and the subjective nature of patient reporting. Electronic health records (EHRs) serve as a digital repository for patient health information. While their primarily provide a wealth of data for monitoring and tracking symptoms, thereby aiding in the management and prevention of a wide range of health conditions and diseases. A significant portion of EHRs is time-stamped, facilitating the longitudinal tracking of a patient’s health status. EHRs have expanded beyond hospital systems to include health diaries collected by smartphone-based applications. These applications allow users to record various lifestyle factors that may influence their health over time. The use of such health apps represents an advancement in healthcare technology, enabling more proactive and personalized approaches to managing health conditions such as migraines. The growing volume of such data presented an opportunity to create comprehensive and practical machine learning models for risk stratification, disease prognosis, and chronic disease management.EHRs have become invaluable tools for tracking various health conditions over time.Researchers have developed effective prediction tools by utilizing deep learning architectures that handle sequential data, such as recurrent neural networks (RNNs), long short-term memory (LSTM) networks, and Gated Recurrent Units (GRUs). These advanced models leverage the temporal nature of EHRs to predict potential future diseases, providing crucial insights for preventative healthcare. Deep learning architectures like recurrent neural networks (RNNs), long short-term memory (LSTM), and Gated Recurrent Units (GRUs) were effective for representing temporal EHRs due to their ability to handle sequential data. These models excel in capturing complex temporal patterns, making them superior choices for various EHR applications. Convolutional neural networks (CNNs), originally designed for image processing, can also be applied to one-dimensional time series data like EHR. They offer parameter efficiency through sparse interactions and parameter sharing as each convolution filter is applied across the entire input. In the literature, a hybrid model combining CNN and RNN (or LSTM) architectures has been proposed for processing temporal EHR data. This hybrid approach leverages the strengths of both architectures: CNNs excel at spatial feature extraction, while RNNs excel at capturing temporal dependencies, making the model well-suited for time- series data. Doctor AI is a well-known example of a model that uses longitudinal time-stamped EHR data to perform multilabel prediction (Choi, Edward, et al. “Doctor AI: Predicting clinical events via recurrent neural networks“). This model was developed using RNN and applied on a cohort of 260K patients over an 8-year period. The main aim of this model was to utilize longitudinal records of patient past hospital visits to predict the diagnosis and the medication prescribed by physicians during the next visit. This model exemplifies the effective application of EHR data in predictive healthcare analytics. In Choi, Edward, et al. “Using recurrent neural network models for early detection of heart failure onset“, researchers investigated the use of deep learning to enhance the prediction of heart failure onset by analyzing EHR data. They used data from 3,884 heart failure cases and 28,903 control subjects, leveraging RNN with GRUmodels to identify temporal relationships among events within a 12- to 18-monthobservation window and predict the likelihood of patient's heart failure. In Stubberud, Anker, et al.: “Forecasting migraine with machine learning based onmobile phone diary and wearable data“, scientists employed mobile phone-basedheadache diaries and wearable devices to automatically monitor potential migraine indicators such as heart rate, peripheral skin temperature, and muscle tension. They aimed to predict migraine occurrences for the following day using various conventional machine learning methods. The top-performing model was Random Forest. It achieved accuracy, recall (sensitivity), and specificity values of 0.56, 0.0,and 1.0, respectively, on a held-out test set. This means the model correctlyidentified all negative cases, achieving perfect specificity, but failed to identify any positive cases. Despite its insights, the study has limitations. Firstly, the sample size was small, involving only 18 patients. Secondly, the study duration was relatively short, lasting 295 days. Consequently, the machine learning models used in the study exhibited a limited generalization capability.In Houle, Timothy T., et al.: “Forecasting individual headache attacks usingperceived stress: development of a multivariable prediction model for persons withepisodic migraine“, researchers developed a multivariable prediction model to forecast headache attacks within 24-hour period based only on stress levels perceived by migraine patients, while overlooking other important migraine trigger factors. Mobile applications dedicated to headache disorders allow users to log numerous headache characteristics, such as start and end times, pain intensity, and whetherthe headache was unilateral or throbbing, if the headache worsened with physicalactivity, was accompanied by nausea, vomiting, light sensitivity (photophobia), sound sensitivity (phonophobia), or aura. Moreover, these apps enable users to monitor lifestyle factors that may influence their headaches, for example, users can record their caffeine intake, fluid consumption, alcohol consumption, missed meals, and other health-related factors such as stress levels, energy levels, sleep duration and quality, and activity levels. Additionally, these apps can gather weather-related data and other sensor information, as discussed in Raffaelli, Bianca, et al.: “Determining the evolution of headache among regular users of a daily electronicdiary via a smartphone app: observational study“. This comprehensive tracking canprovide valuable insights for both users and healthcare professionals, facilitating more effective management and treatment of headache disorders.It is therefore an objective of the present invention to improve and further develop amethod, system and computer program for predicting occurrence of a recurring medical symptom or body behavior, such as a migraine attack.In accordance with the invention, the aforementioned objective is addressed by acomputer-implemented method comprising the features of claim 1. According to this claim, such a method for optimizing a prediction of an occurrence of a recurring medical symptom or body behavior of a human being using machine learningcomprises obtaining input data comprising features related to the occurrence of therecurring medical symptom or body behavior. For example, the input data may be derived from or include data of an electronic health record. For example, the input data may be generated by a mobile application being executed on a mobile device, such as a smartphone, smartwatch or other type of (medical) wearable device. Alternatively, the input data may be generated by an application running on a server and be derived from data provided by the human being (and its mobile device(s)). The method comprises dividing the features included in the input data into two or more groups of features. The method comprises encoding, using a trained encoder machine learning model, the features of the two or more groups of features into twoor more embeddings. Each embedding represents a group of features. The methodcomprises inputting the two or more embeddings into a prediction machine learningmodel being trained to predict the occurrence of a recurring medical symptom or body behavior based on the two or more embeddings. The method comprises providing a prediction of the recurring medical symptom or body behavior based on an output of the prediction machine learning model. The proposed concept is based on the finding, that, by grouping the features into groups of features (also denoted “concepts” in the present disclosure) that are condensed in a lower-dimensional embedding, additional context is encoded into the representations of the respective features, with a focus on features that are relevant for the subsequent prediction of the occurrence of the recurring medical symptom or body behavior. For example, the technique outlined in Koh, Pang Wei, et al. “Concept bottleneck models” may be used for this purpose. This way, less relevant features are deemphasized in the respective context, which facilitates the training and improves the predictive capabilities of the prediction machine learningmodel being used to make the prediction. The predictive capabilities of the machinelearning pipeline (comprising the encoder machine learning model and the prediction machine learning model) can be applied to recurring medical symptoms, such as migraine attacks, as well as to the analysis of features related to sleep, i.e. to determine which features or groups of features primarily impact the sleep of the human being.To further improve the predictive capabilities of the machine learning pipeline, thetemporal context of the input data can be taken into account. For example, the input data may be longitudinal input data comprising a plurality of observations of the same subject over a plurality of points in time. In this context, longitudinal data, also known as panel data, is a type of data that involves observations of the same subjects (such as individuals, groups, or organizations) repeatedly over a period oftime. This longitudinal data is represented in the embeddings being generated, bygenerating embeddings that each represent multiple time steps. Thus, the two ormore embeddings may be multi time-step embeddings each representing a group of features over multiple points in time. By providing the additional temporal context, the predictive performance of the machine learning pipeline can be improved further.For example, the two or more multi time-step embeddings may each comprise aconcatenation of single time-step embeddings output by the trained encoder machine learning model, with each single time-step embedding representing a point of time. This facilitates generation of the multi time-step embeddings as part of the machine learning pipeline, as the single time-step embeddings can be used across multiple time-steps. In general, each of the multi time-step embeddings may represent a window of time that shifts according to the time step for which the prediction is to be performed.Thus, the two or more multi time-step embeddings may be encoded based on awindow of points of time (i.e., a subset of points of time that is selected according toa window of time) within the plurality of points in time.In addition to the time-step concatenation (or as an alternative), the embeddings may also be concatenated across groups of features. In other words, the method may comprise concatenating the two or more embeddings to obtain a concatenated embedding representing the groups of features and providing the concatenated embedding as input for the prediction machine learning model. By using a concatenated embedding representing the groups of features, a single, time-aligned input can be provided to the prediction machine learning model, ensuring that the prediction is based on the same point(s) in time of input data. The proposed concept is based on grouping together features into groups (also denoted “concepts”). These high-level groups or concepts can be inferred using a clustering algorithm (e.g., K-means clustering, hierarchical clustering, Gaussian Mixture Models) or specified by domain experts to allow for more comprehensive representation learning. In particular, features may be grouped together that are topically related, e.g., as shown in Fig.7, where the features are grouped into the concepts “demographics” (including gender and age at signup), “behavioral” (smoking, alcohol consumption, caffeine consumption, fluid intake, missed meals), “health factors” (sleep duration, sleep quality, stress, activity, energy, mood), “immutable” (air pressure, humidity, temperature, weekday, season) and “medication” (medication, medical class, non-opioid analgesic, triptane, betablocker, magnesium, anti-epileptic, combination analgesic, anti-depressant,calcium antagonist, CGRP (calcitonin gene-related peptide) blocker). Thus, the twoor more groups of features may comprise at least one group comprising topicallyrelated features. For example, the two or more groups of features may comprise at least one of a group of features related to a behavior of the human being, a group of features related to a demographic of the human being, a group of features related to health features of the human being, a group of features related to medication taken by the human being and a group of features related to environmental factors (“immutables”). Such grouping can improve the predictive capabilities of the machine learning pipeline. In addition to the prediction of the recurring medical symptom or body behavior, another output of the machine learning pipeline may include an explanation of the prediction, i.e., an explanation as to which feature has an increased impact on the prediction. This can be done using explainable artificial intelligence (AI) techniques,such as SHAP (SHapley Additive exPlanations) or LIME (Local Interpretable Model-agnostic Explanations). In particular, the method may comprise providing, based on a set of Shapley values, information on an impact of features of the features related to the occurrence of the recurring medical symptom or body behavior on the prediction. This may help the human being understand what the likely cause of a predicted recurrence of the medical symptom is, e.g., of the migraine attack. This way, the human being can make minor adjustments in their behavior or medication based on the explanation to avoid recurrence of the recurring medical symptom. There are various ways for implementing the respective machine learning models. For example, the encoder machine learning model may be implemented using oneof a Long Short-Term Memory (LSTM), a Convolutional Neural Network (CNN), anda Long Short-Term Memory-Convolutional Neural Network hybrid. By using a temporally aware machine learning model, such as an LSTM or an LSTM hybrid, the temporal information contained in the input data can be used as additional context when encoding the groups of features. For example, the prediction machine learning model may be implemented using one of a random forest, a support vector machine and a neural network. By inputting the features as embedding into the prediction machine learning model, simpler model types, such as random forests or support vector machines, or relatively simple neural networks can be used as predictor, which can decrease the effort for training the prediction machine learning model and for performing inference using the prediction machine learning model. Further aspects of the present disclosure relate to a corresponding computer program, non-transitory, computer-readable medium comprising a program code and system for performing the above method. For example, an aspect relates to a computer program comprising instructions which, when the program may be executed by a computer, cause the computer to carry out the above method. Another aspect relates to a non-transitory, computer-readable medium comprising a program code that, when the program code may be executed on a processor, a computer, or a programmable hardware component, causes the processor, computer, or programmable hardware component to perform the above method. Moreover, another aspect relates to a system comprising interface circuitry, machine-readable instructions, and processor circuitry to execute the machine- readable instructions to carry out the above method.Another set of aspects of the present disclosure relate to a computer-implementedmethod, computer program, program code and system for training at least onemachine learning model for use in predicting occurrence of a recurring medicalsymptom or body behavior, e.g., of a target human being. The method comprisesobtaining a set of training data. The set of training data comprises a plurality ofsamples of input data. As outlined above, the samples of input data may compriseor be derived from electronic health records and may be provided by a mobileapplication or server application. Each sample of input data comprises featuresrelated to the occurrence of the recurring medical symptom or body behavior. The method comprises dividing the features included in the samples of input data into two or more groups of features. The method comprises training an autoencoder comprising an encoder machine learning model and a decoder machine learning model using the set of training data. The autoencoder is trained to encode, using the encoder machine learning model, the respective groups of features into anembedding representing the respective group of features. The resulting embeddingshave a lower dimensionality than the respective groups of features represented bythe embeddings. The autoencoder is trained to decode, using the decoder machinelearning model, the embeddings to restore the groups of features. The method comprises providing the encoder machine learning model of the autoencoder for use in encoding the features of the two or more groups of features into embeddings. By grouping the features into groups of features (also denoted “concepts” in the present disclosure) that are condensed in a lower-dimensional embedding, additional context is encoded into the representations of the respective features,with a focus on features that are relevant for the subsequent prediction of theoccurrence of the recurring medical symptom or body behavior. This way, less relevant features are deemphasized in the respective context, which facilitates the training and improves the predictive capabilities of a prediction machine learning model being used in combination with the encoder machine learning model to make the prediction. The predictive capabilities of the entire machine learning pipeline (comprising the encoder machine learning model and the prediction machine learning model) can be applied to recurring medical symptoms, such as migraine attacks, as well as to the analysis of features related to sleep, i.e. to determine which features or groups of features primarily impact the sleep of the human being. The proposed concept is based on applying machine learning to generate embeddings and use the embeddings to predict occurrence of a recurring medical symptom or body behavior. Machine learning is a branch of artificial intelligence that involves the development of algorithms and models that allow computers to learn and make predictions or decisions without being explicitly programmed. It focuseson creating systems that can improve their performance over time by learning fromdata. Training a machine-learning model refers to the process of teaching the model to make accurate predictions or decisions. During training, the model is exposed to a large amount of data, which is used to adjust the model's internal parameters or weights. The model learns patterns, relationships, or rules from the training data, allowing it to generalize and make predictions on new, unseen data. Training data is the set of examples or instances that is used to teach a machine- learning model. It is often labeled data, meaning that each example is associated with a known outcome or target value. The training data consists of both input features and the corresponding output or target variable. The model learns from this data by analyzing the patterns and relationships between the input features and the target variable. Training algorithms, such as supervised learning, semi-supervised learning, unsupervised learning or reinforcement learning may be used for training the machine-learning model.In the present concept, three machine learning models are used – an encodermachine learning model, a prediction machine learning model, and a decoder machine learning model (that is only used during training of the encoder machine learning model). Machine learning models are trained using machine learning. In particular, the autoencoder, and thus the encoder machine learning model and the decoder machine learning model, may be trained using unsupervised learning. Anautoencoder is a type of machine learning model (e.g., neural network) that learnsto copy its input to its output using unsupervised learning. It comprises two main parts: the encoder and the decoder. The encoder compresses the input data into alatent-space representation or encoded form (i.e., into the embeddings), and thedecoder attempts to reconstruct the input data from this encoded form. The trainingof an autoencoder is unsupervised because it does not require any labels or targetdata; it essentially learns from the input data itself. To train the autoencoder, thesamples of input data are input into the encoder machine learning model, whichcompresses it. This compressed data is passed through the decoder machine learning model which then attempts to recreate the original input data from this compressed form. The training process involves minimizing a loss function thatmeasures the difference between the output of the decoder machine learning model(the reconstructed inputs) and the original inputs. By minimizing this loss, the autoencoder learns efficient data codings in an unsupervised manner. The resulting encodings capture the most salient features of the data necessary to reconstruct the input as closely as possible, thus learning a compressed knowledge representation of the data. The other model being used by the method being used to predict occurrence of the recurring medical symptom or body behavior is the prediction machine learning model. Accordingly, the method may comprise training a prediction machine learning model to predict the occurrence of the recurring medical symptom or body behavior based on the embeddings output by the encoder machine learning model. In particular, the prediction encoder model may be trained using supervised learning. For this purpose, the training data may comprise, for each or a subset of training samples, a desired output for the prediction machine learning model. For example, the label may be a binary label indicating whether the recurring medical symptom or body behavior (e.g., the migraine attack) has occurred at a given point in time relative to the point(s) in time the input data samples represent. Thus, the prediction machine learning model may be trained as a classifier, with the output representing whether the recurring medical symptom or body behavior is predicted to occur at the given point in time relative to the input data being fed into the prediction machine learning model (and the encoder machine learning model, toprepare the embeddings that are used as input for the prediction machine learningmodel). The method may further comprise providing the prediction machine learningmodel together with the encoder machine learning model. The encoder machine learning model and the prediction machine learning model are used together as part of a machine learning pipeline. In particular, input data (e.g., the input data samples of the training data during training, or new input data during production use) is input into the encoder machine learning model, which outputs embeddings representing the input data. The resulting embeddings are input into the prediction machine learning model, resulting in a machine learning pipeline comprising both the encoder machine learning model and the prediction machine learning model. Between the two machine learning models, some data processing may be performed to adapt the raw embeddings output by the encoder machine learning model for the prediction machine learning model. In particular, each sample of input data may be longitudinal input data comprising a plurality of observations of the same subject over a plurality of points in time. These points in time may be (partially) represented by the embeddings. For example, the encodermachine learning model may be used to output, for each point in time represented by the input data and for each group of features, an embedding. These embeddings may be concatenated, using a windowing approach and by group, into multi time- step embeddings. Thus, training the prediction machine learning model may comprise determining two or more multi time-step embeddings each representing agroup of features over multiple points in time (by concatenating the single time-stepembeddings into the multi time-step embeddings). For example, the two or moremulti time-step embeddings may be encoded based on a window of points of timewithin the points in time represented by the respective input data sample. For example, the method may comprise concatenating the embeddings output by the encoder machine learning model to obtain the two or more multi time-step embeddings. The two or more multi time-step embeddings may then be input into the prediction machine learning model. Moreover, concatenation might not only be performed across time-steps, but alsoacross groups. For example, the method may comprise concatenating the two ormore multi time-step embeddings to generate a concatenated embedding, with the concatenated embedding being input into the prediction machine learning model. Machine-learning models, such as the machine-learning model being trained in the present disclosure, are often implemented as Artificial Neural Networks (ANNs), and in particular Deep Neural Networks, Support Vector Machines, Decision Treemodels, or Random Forest models. In particular, as outlined above, the encodermachine learning model may be implemented using one of a LSTM, a CNN, and aLSTM-CNN hybrid. The prediction machine learning model may be implementedusing one of a random forest, a support vector machine and a neural network. The best prediction performance may be achieved by tailoring the machine learning pipeline to a specific user (a target human being), i.e., to the user that is to use the machine learning pipeline. This can be done by using input data that is provided by the target human being, e.g., through use of a mobile application being executed by a mobile device of the target human being. Thus, the encoder machine learningmodel and / or the prediction machine learning model may be at least partially trained(e.g., fine-tuned) based on input data samples related to a target human being. Togive a suitable starting point for fine-tuning, and to obtain models that are better at generalization, the models may be pre-trained using the input data of other humanbeings. In other words, the encoder machine learning model and / or the predictionmachine learning model may be at least partially trained based on input data samples related to a plurality of human beings. One core aspect of the proposed concept lies in the division of the features into groups. In particular, the features are divided into the two or more groups of features according to a mapping between features and groups of features. This mapping may be determined, e.g., as part of the training, using a clustering algorithm. Forexample, a contextual analysis module may assess the relevance and relationshipsof features based on the data, application context, and expert knowledge, to groupfeatures into groups (concepts) and allow the integration of new concepts as newfeatures are introduced. For example, the groups may be dynamically adjusted. Inparticular, machine learning clustering techniques may be used to dynamicallyadjust the concept groups. For instance, if certain concepts become more relevantor change their relationships, the grouping algorithm may update the mapping, andthus the groupings, accordingly. In an alternative implementation, the mappingbetween features and groups of features may be pre-defined according to expert knowledge. In addition to the prediction of the recurring medical symptom or body behavior, another output of the machine learning pipeline may include the explanation of the prediction, i.e., an explanation as to which feature has an increased impact on the prediction. This can be done using explainable artificial intelligence (AI) techniques,such as SHAP (SHapley Additive exPlanations) or LIME (Local Interpretable Model-agnostic Explanations). In the context of the training method, the method may comprise computing, using the encoder machine learning model and the prediction machine learning model, a set of Shapley values. The set of Shapley valuesrepresents an impact of features of the features related to the occurrence of therecurring medical symptom or body behavior on a prediction provided by theprediction machine learning model. This set of Shapley values may be used, as partof an inference operation, to provide an explanation of the prediction being provided by the machine learning pipeline. This may help the human being understand what the likely cause of a predicted recurrence of the medical symptom is, e.g., of the migraine attack. This way, the human being can make minor adjustments in their behavior or medication based on the explanation to avoid recurrence of the recurring medical symptom. Further aspects of the present disclosure relate to a corresponding computer program, non-transitory, computer-readable medium comprising a program code and system for performing the above method. For example, an aspect relates to a computer program comprising instructions which, when the program may be executed by a computer, cause the computer to carry out the above method. Another aspect relates to a non-transitory, computer-readable medium comprising a program code that, when the program code may be executed on a processor, a computer, or a programmable hardware component, causes the processor, computer, or programmable hardware component to perform the above method. Moreover, another aspect relates to a system comprising interface circuitry, machine-readable instructions, and processor circuitry to execute the machine- readable instructions to carry out the above method. There are several ways how to design and further develop the teaching of the present invention in an advantageous way. To this end it is to be referred to the dependent claims on the one hand and to the following explanation of preferred embodiments of the invention by way of example, illustrated by the figure on the other hand. In connection with the explanation of the preferred embodiments of the invention by the aid of the figure, generally preferred embodiments and further developments of the teaching will be explained. In the drawingFig. 1a shows a flow chart of a computer-implemented method foroptimizing a prediction of an occurrence of a recurring medical symptom or body behavior of a human being using machine learning;Fig. 1b shows a block diagram of a system for optimizing a prediction of anoccurrence of a recurring medical symptom or body behavior of a human being using machine learning;Fig. 2a shows a flow chart of a computer-implemented method for trainingat least one machine learning model for use in predicting occurrence of a recurring medical symptom or body behavior;Fig. 2b shows a block diagram of a system for training at least one machinelearning model for use in predicting occurrence of a recurring medical symptom or body behavior;Figs. 3a to 3d give an overview over a dataset;Fig. 4 shows an overview of a training portion of a Temporal RecurrentAutoencoder for Concept Embedding (TRACE) algorithm;Fig. 5 shows an overview of an integration of TRACE with a trainedclassifier;Fig. 6 provides a detailed outline of an example implementation of theTRACE algorithm;Fig. 7 shows an example feature list;Fig. 8a and 8b compare a per-class recall across five distinct runs for competingmethods using 2-day forecast and 3-day forecast datasets;Figs. 9a to 9c present a comparative analysis of three key performance metrics:precision, recall, and macro F1 score;Fig. 10 shows the corresponding weighted key performance metrics:precision, recall, and F1 score values;Fig. 11 summarizes an average training time in minutes for each of thecompeting model and the standard deviation in minutes across 5 runs for each mode;Fig. 12 provides a global overview of feature importance derived from aRandom Forest model leveraged in the TRACE algorithm;Fig. 13 shows an ablation analysis illustrating the recall per class forindividual feature concepts and their combination using the 3-day forecast dataset;Fig. 14 shows precision, recall, macro F1 score, and the correspondingweighted values across five different runs for individual feature concepts and their combination using the 3-day forecast dataset;Fig. 15a to 15j display SHAP summary plots of the top 10 features influencingmigraine occurrences for users who consistently used the application over an extended period; andFig. 16 shows another illustration of an analysis of feature importance for aspecific user. Embodiments of the present disclosure relate to a method, system and computerprogram for optimizing a prediction of an occurrence of a recurring medical symptomor body behavior of a human being using machine learning, and to a method, system and computer program for training the machine learning models being used by such a method, system and computer program. The method, system and computer program for optimizing a prediction of an occurrence of a recurring medical symptom or body behavior of a human being using machine learning is discussed in connection with Figs.1a and 1b, while the method, system and computer program for training the machine learning models is discussed in connection with Figs. 2a and 2b. Figs.3a to 15 show an example implementation, denoted TRACE, of the methods, systems and computer programs discussed in connection with Figs.1a to 2b. Fig. 1a shows a flow chart of a computer-implemented method for optimizing a prediction of an occurrence of a recurring medical symptom or body behavior of a human being using machine learning. The method comprises obtaining 110 input data comprising features related to the occurrence of the recurring medical symptom or body behavior. The method comprises dividing 120 the features included in the input data into two or more groups of features. The method comprises encoding 130, using a trained encoder machine learning model, the features of the two or more groups of features into two or more embeddings, with each embeddingrepresenting a group of features. Optionally, the method comprises concatenating140 the two or more embeddings to obtain a concatenated embedding representing the groups of features. The method comprises inputting 150 the two or moreembeddings (e.g., their concatenation) into a prediction machine learning modelbeing trained to predict the occurrence of a recurring medical symptom or body behavior based on the two or more embeddings. The method comprises providing 160 a prediction of the recurring medical symptom or body behavior based on an output of the prediction machine learning model. Optionally, the method further comprises the method comprises providing 170, based on a set of Shapley values and based on the input data, information on an impact of features of the features related to the occurrence of the recurring medical symptom or body behavior on the prediction. The method of Fig.1a may be performed by the corresponding system 10 shown in Fig.1b. Fig.1b shows a block diagram of a system 10 for optimizing a prediction of an occurrence of a recurring medical symptom or body behavior of a human being using machine learning. The system 10 shown in Fig. 1b is a computer system comprising interface circuitry 12, processor circuitry 14, machine-readableinstructions, and, optionally, memory and / or storage circuitry 16. The interfacecircuitry 12 may be used to facilitate communication with other components, such as sensors, or with systems of computers. This communication may occur via a bus or a network, for example. The processor circuitry 14 may be used to provide thefunctionality of the system, for example, in conjunction with the interface circuitry 12(for exchanging information) and / or optional memory or storage circuitry 16 (for storing information, such as machine-readable instructions). The processor circuitry 14 is therefore coupled with the interface circuitry 12 and, optionally, with the memory or storage circuitry 16. For instance, the system 10 may comprise machine- readable instructions, i.e., a computer program, which may prompt the one or more processors to execute at least one of the methods introduced in connection with Fig. 1a. In accordance with the aforementioned, the method introduced in connection with Fig. 1a may be carried out by the one or more processors executing the machine-readable instructions. For example, the interface circuitry 12 may include or correspond to a network interface circuitry and / or a device interface circuitry configured to be communicatively coupled to one or more other devices, such as the one or more processors. For example, the interface circuitry may include a transmitter, a receiver, or a combination thereof (e.g., a transceiver), and may enable wired communication, wireless communication, or a combination thereof. For example, the processor circuitry 14 may include or correspond to one or more of a digital signal processor circuitry (DSP), a graphical processing unit (GPU), and / or a central processing unit (CPU). For example, the memory and / or storage circuitry 16 may include or correspond to volatile or nonvolatile storage circuitry, such as Random Access Memory (RAM), magnetic disks, optical disks, or flash memory devices. Theone or more memory / storage devices may include both removable and non-removable memory devices. Fig.2a shows a flow chart of a computer-implemented method for training at least one machine learning model for use in predicting occurrence of a recurring medicalsymptom or body behavior, e.g., of a target human being, e.g., for use by the methodand system discussed in connection with Figs.1a and 1b. The method comprises obtaining 210 a set of training data. The set of training data comprises a plurality of samples of input data. Each sample of input data comprises features related to the occurrence of the recurring medical symptom or body behavior. The methodcomprises dividing 230, according to a mapping between features and groups offeatures, the features included in the samples of input data into two or more groups of features. For example, the mapping between features and groups of features may be pre-defined according to expert knowledge, or may be determined 220 using a clustering algorithm. The method comprises training 240 an autoencoder comprising an encoder machine learning model and a decoder machine learning model using the set of training data. The autoencoder is trained to encode, using the encoder machine learning model, the respective groups of features into an embedding representing the respective group of features. The resulting embeddings have a lower dimensionality than the respective groups of features represented by the embedding, and to decode, using the decoder machine learning model, the embeddings to restore the groups of features. Optionally, the method comprises training 260 a prediction machine learning model to predict the occurrence of therecurring medical symptom or body behavior based on the embeddings output bythe encoder machine learning model. For example, the embeddings output by the encoder machine learning model may be concatenated 250 before they are input into the prediction machine learning model. The method comprises providing 270 a) the encoder machine learning model of the autoencoder for use in encoding the features of the two or more groups of features into embeddings, and b) (optionally) the prediction machine learning model. Further optionally, the method comprises computing and providing 280, using the encoder machine learning model and theprediction machine learning model, a set of Shapley values, with the set of Shapleyvalues representing an impact of features of the features related to the occurrence of the recurring medical symptom or body behavior on a prediction provided by the prediction machine learning model. The method of Fig.2a may be performed by the corresponding system 20 shown in Fig.2b. Fig.2b shows a block diagram of a system for training at least one machine learning model for use in predicting occurrence of a recurring medical symptom or body behavior. The system 20 shown in Fig.2b is a computer system comprising interface circuitry 22, processor circuitry 24, machine-readable instructions, and,optionally, memory and / or storage circuitry 26. The interface circuitry 22 may beused to facilitate communication with other components, such as sensors, or with systems of computers. This communication may occur via a bus or a network, for example. The processor circuitry 24 may be used to provide the functionality of the system, for example, in conjunction with the interface circuitry 22 (for exchanging information) and / or optional memory or storage circuitry 26 (for storing information, such as machine-readable instructions). The processor circuitry 24 is therefore coupled with the interface circuitry 22 and, optionally, with the memory or storage circuitry 26. For instance, the system 20 may comprise machine-readable instructions, i.e., a computer program, which may prompt the one or more processors to execute at least one of the methods introduced in connection with Fig. 2a. In accordance with the aforementioned, the method introduced in connection with Fig. 2a may be carried out by the one or more processors executing the machine-readable instructions. For example, the interface circuitry 22 may include or correspond to a network interface circuitry and / or a device interface circuitry configured to be communicatively coupled to one or more other devices, such as the one or more processors. For example, the interface circuitry may include a transmitter, a receiver, or a combination thereof (e.g., a transceiver), and may enable wired communication, wireless communication, or a combination thereof. For example, the processor circuitry 24 may include or correspond to one or more of a digitalsignal processor circuitry (DSP), a graphical processing unit (GPU), and / or a centralprocessing unit (CPU). For example, the memory and / or storage circuitry 26 may include or correspond to volatile or nonvolatile storage circuitry, such as Random Access Memory (RAM), magnetic disks, optical disks, or flash memory devices. Theone or more memory / storage devices may include both removable and non-removable memory devices. In the following, an example implementation of the methods, systems and computerprograms of Figs. 1a to 2b is provided.In the present disclosure, an AI-powered system for forecasting migraine attacksand identifying their triggers is presented. Forecasting migraine attacks before theyoccur can significantly enhance the quality of life for sufferers by allowing them totake preemptive measures to mitigate or prevent the onset of symptoms. In thefollowing, a technique is discussed that aims to predict migraine attacks, with anexample implementation that leverages a comprehensive migraine dataset collected from a mobile app between 2016 and 2022, encompassing approximately 43,000 users and 7 million daily records. In the present disclosure, TRACE(Temporal Recurrent Autoencoder for Concept Embedding) is introduced, a novelapproach developed to anticipate and forecast migraine attacks before they occurusing mobile app-based headache diaries and / or electronic health records. Theproposed method leverages diverse headache and life style features relevant tomigraine attacks (e.g., the features related to the occurrence of the recurringmedical symptom or body behavior). It begins by generating feature concepts, i.e.,relevant feature groups (see Koh, Pang Wei, et al. "Concept bottleneck models"),based on timestamps. Then, a (deep learning-based) autoencoder (e.g., theencoder machine learning model) may be used to generate lower-dimensionalembeddings (e.g., the two or more embeddings) from these timestamped concepts.This strategy aims at facilitating a deeper understanding of migraine patterns and potential triggers by considering both the feature concepts and their timing. Theseembeddings may be subsequently input into a simpler model (e.g., the predictionmachine learning model) such as Random Forest to predict the occurrence of amigraine attack the following day. Additionally, TRACE may leverage advancedExplainable AI (XAI) techniques to identify both individual-specific and global migraine triggers across diverse user populations. On two held-out sets for 2-day and 3-day forecasts, TRACE successfully predicts approximately 70% of migraine episodes before their onset, outperforming all competing methods while maintaining comparable precision and F1 score. To understand the predictive factors driving model performance, both global and localfeature importance analyses are conducted. Global importance is assessed usingablation analysis and Random Forest, while local importance is evaluated on a per-patient basis using SHapley Additive exPlanations (SHAP). The SHAP analysisshows that while common migraine triggers exist, individual triggers can varysignificantly, highlighting the need for personalized predictive tools. The presenttechnique relates to health Informatics, deep learning, electronic health records,autoencoder, and explainable AI.TRACE begins by generating feature concepts (e.g., the two or more groups offeatures), i.e., relevant feature groups, based on timestamps. An (LSTM-based)autoencoder (e.g., the encoder machine learning model) then generates lower-dimensional embeddings from these timestamped concepts. This strategy aims at facilitating a deeper understanding of migraine patterns and potential triggers by considering both the feature concepts and their timing. These embeddings aresubsequently input into a (Random Forest) model (e.g., the prediction machinelearning model) to predict the occurrence of a migraine attack the following day.TRACE is evaluated against several established methods, including LSTM (LongShort-Term Memory), CNN (Convolutional Neural Network), a hybrid LSTM-CNNmodel, and a stand-alone Random Forest. Additionally, an LSTM model withSHapley Additive exPlanations (SHAP) is leveraged to gain further insights into patient-specific migraine triggers.The example implementation discussed herein uses the M-sense dataset. M-sensewas a smartphone-based electronic headache diary administered by patients. This mobile app was available in Germany, Austria, and Switzerland for Android from2016 and iOS from 2017 until 2022. With M-sense, users logged different featuresof their headaches, such as the start and end times, the maximum intensity of pain (rated on a scale from 0 to 10), and binary features like whether the headache was unilateral, throbbing, worsened with physical activity, accompanied by nausea or vomiting, and whether there was sensitivity to light (photophobia), sound (phonophobia), or if they experienced migraine aura. Using these headache features, M-sense classified each headache occurrence as migraine, tension-type headache (TTH), or non-headache using an algorithm that adheres to the International Classification of Headache Disorder guidelines (ICHD-3). M-sense users were also able to document prescribed or administered medications, including the specific name and the time it was taken. They could log their medications, irrespective of whether they were prescribed for migraines or other chronic or acute conditions. Additionally, M-sense enabled users to track various lifestyle factors that might have influenced their headaches. Users could record their caffeine consumption, fluid intake, alcohol consumption, missed meals, as well as other health-related factors such as stress levels, energy levels, sleep duration, sleep quality, activity levels, and mood. For health-related factors, the app user is requested to assign numerical values ranging from 1 to 10, with 1 being the minimum and 10 being the maximum. Additionally, the app automatically gathered weather data, i.e., temperature, humidity, and air pressure. M-sense stored demographic data, namely, age and gender. Many of these features have been shown in previous studies to potentially trigger or contribute to prolonged headache attacks. In a prior publication, a subset of the M-sense data, containing 1545 users, wasused in an observational study aimed at tracking the progression of headachesamong regular M-sense users. A larger subset of the same dataset, which included 2325 users, was leveraged to investigate the prevalence of headaches during the COVID-19 lockdown in Germany. The entire dataset collected by M-sense from2016 to 2022 was acquired, which includes data from approximately 43,000 userswith 7 million daily records. This dataset is composed of structured data and does not include any unstructured entries such as user-input diaries. For the purpose of the example implementation of the proposed technique, dailyrecords labeled as either migraine or no-headache instances were extracted,excluding tension-type headaches (TTH) since TTH accounted for less than 5% ofthe records. To ensure the integrity and completeness of the dataset, a stringentfilter was applied to retain only those records that exhibited no missing values acrosssix key features. These features, which are referred to in the following as ‘dailyhealth factors’, include sleep duration, sleep quality, energy level, mood, stress, and activity. Imputing missing values for these features posed a considerable challenge.All users who experienced at least a single migraine attack while using the app wereincluded, as the goal is to forecast migraines and identify their triggers. Headache-related features were excluded from the predictive modeling task because thesefeatures were used to determine the label of each daily record, which is categorized as either ’migraine’ or ’no-headache’. In summary, abstracting from the M-sense dataset, the proposed concept leverages comprehensive headache datasets that encompass longitudinal data such asheadache features, lifestyle factors, health variables, sensor data, and medical dataand medical imaging collected over different time points for various users who suffered from migraine.Feature pre-processing was conducted on the dataset, including meannormalization per user for features such as sleep duration and quality, alcoholconsumption, smoking count and caffeine intake. One-hot encoding was applied tofeatures such as dataset comprises 3,819,618 daily records from 22,619 users. The number of daily records labeled as ‘migraine’ is 554,625, while the number of daily records labeled as ‘no headache’ is 3,264,993. Therefore, the approximate ratio of records labeled ‘migraine’ to those labeled ‘no headache’ is 1:6. Figs.3a to 3d give an overview over the dataset. Fig. 3a shows the distribution of users by gender.Migraine prevalence is higher among females compared to males in the dataset.Fig.3b shows daily record counts, grouped by label and gender. Fig.3c shows agedistribution. The majority of migraine patients fall between the ages of 22 and 55.Fig. 3d shows the average number of migraine days per month experienced byusers. The majority of users in the dataset reported having fewer than ten migrainedays per month.Assuming a set of N migraine patients; each patient, pi, is observed at Ti differenttime points, and each observation comprises or consists of F features. This data can be viewed as a set of multivariate time series, where each pi is represented by^^ ∈ ℝ^^×^. Here, the following notation is used: Xi,t,f is the value of the f-th feature for the i-th patient at the t-th time point, Xi,t is the feature vector for the i-th patient at the t-th time point, and yi,t is the label for the i-th patient at the t-th time point.For the example implementation and technique, daily records were considered,where each record comprises or consists of various features collected over a single day for an individual patient. To prepare time series datasets for forecastingmigraine attacks, samples were created by sliding a window of size w, andextracting sequences of length w days. In more detail, for each patient pi, asequence was extracted at each time ^ ∶ ^ < ^ ≤ ^^, such that:- a sequence at time t is extracted only if there is no headache occurrence atthe previous w days, i.e., ^^,^ = 0 for all ^ ∈ {^ − ^, ... , ^ − 1}.- The sequence is built from the feature vector from theprevious w days. -Extract label ^^,^ for the current day ^.^"migraine" if ^^,^is a migraine event "^^ ℎ^^^^^ℎ^" ^^ℎ^^^^^^- Form training samples as pairs (^^,(^^^:^^^), ^^,^). Here, representsthe feature matrix of the days preceding the current day t. These features predict whether or not the user pi will experience a migraine attack on the day t, as indicated by the label yi,t. -Ni represents the number of valid extracted time series sequences, i.e., patient pi.Two datasets were retrieved corresponding to w = 2 and w = 3, referred to as the 2-day and 3-day forecasts, respectively. Each dataset can be represented by:^ is the number of valid sequences for all users.Initially, the Temporal Recurrent Autoencoder for Concept Embedding (TRACE)method may build upon the structure of a traditional autoencoder. An autoencoderis a type of neural network designed to compress input data ^ into a lower-dimensional representation, denoted as ^, and then reconstruct an approximation of the original data, ^^, from ^. During a training phase, the autoencoder aims toapproximate the identity function, ℎ^,^(^) ≈ ^, by setting the target values to beequal to the inputs. Through unsupervised learning, autoencoders may thus identify latent variables that influence the underlying data distribution. These latent variables, collectively known as the latent space, represent the most important information in the input data. The autoencoder learns to use these variables to accurately reconstruct the original data. The learning process involves backpropagation, where the target values are set to be equal to the inputs, makingthe autoencoder learn an approximation to the identity function, ℎ^,^(^) ≈ ^, whereW represents the weight matrix and b represents the bias vector of the neuralnetwork. By imposing constraints on the network, such as limiting the number ofhidden units, interesting structures within the data can be uncovered (seeGoodfellow et al.: Deep Learning). TRACE enhances the capabilities of a standard autoencoder by generatingembeddings for feature concepts, not raw features, and by taking into considerationthe temporal nature of input data. In the proposed method, features are groupedinto relevant high-level groups (i.e., concepts), which can be inferred using a clustering algorithm (e.g., K-means clustering, hierarchical clustering, GaussianMixture Models) or specified by domain experts to allow for more comprehensiverepresentation learning (see Koh, Prang Wei et al.: “Concept Bottleneck Models”). For example, features such as air pressure, humidity, temperature, day of the week, and season are grouped under the “weather” concept, while features like smoking, alcohol consumption, caffeine intake, and missed meals fall under the “behavioral” concept. Additionally, TRACE incorporates a time-step aspect where each time-step (i.e., a fixed unit of time, such as an hour, a day or a month) is encoded separately then their representation is concatenated to generate ^. By leveraging this structure,TRACE can effectively compress sequential data into a lower-dimensional spacewhile preserving its temporal characteristics. This makes TRACE particularly well-suited for tasks such as time-series forecasting (see Fig. 4).With respect to the methods shown in Figs.1a and 1b, this means that the input data used therein may be longitudinal input data comprising a plurality of observations of the same subject over a plurality of points in time. The two or more embeddings being generated with help of the encoder machine learning models multi time-step embeddings each representing a group of features over multiple points in time. These can be obtained by concatenating the single time-step embeddings output by the trained encoder machine learning model. Further optionally, the multi time-step embeddings may be concatenated, such that the concatenated embedding represents multiple groups of features, over multiple points in time.Fig. 4 shows an overview of a training portion of TRACE. Fig. 4 may provide anexample implementation of the method of Fig.2a. F refers to the total number of features, N refers to the number of samples, and w refers to the observation window(i.e., total number of time points). In Fig. 4, first, the proposed method groupsfeatures into relevant high-level concepts (groups of features), i.e., ^^, ^^, … , ^^.Second, it splits each concept by time-step (with each time-step representing a pointin time). For simplicity, the figure illustrates an example with three-time steps denoted as ^^, and ^^. Third, the encoder converts time-step concepts into lower-dimensional representations, ^^, ^^, … , ^^ (the embeddings). Then, using the latentspace, the decoder follows the same steps as the encoder but in reverse to reconstruct an image of the original data. The learning process involves backpropagation, where the target values are set to be equal to the inputs, makingthe autoencoder learn an approximation to the identity function, ℎ^,^(^) ≈ ^,where ^ represents the weight matrix and ^ represents the bias vector of the neuralnetwork. By imposing constraints on the network, such as limiting the number of hidden units, interesting structures can be uncovered within the data. As an example both the encoder and decoder can be implemented using LSTM networks or CNN or a hybrid model of LSTM-CNN.In Fig. 5, which provides an example implementation of the method of Fig. 1a, theproposed method integrates the trained encoders (e.g., the encoder machinelearning model) with a classifier (e.g., the prediction machine learning model). Arandom forest model is used as an example; however, it can be replaced with othermodels depending on the requirements and complexity of the input data, the randomforest can be replaced with a simpler model like Support Vector Machine (SVM) ormore complex models like neural networks. The trained encoders are leveraged to generate meaningful representations that capture important features and temporal dependencies from the input concepts. These learned representations for each concept are then used as inputs to Random Forest for predicting future migraine attacks.Fig. 6 provides a detailed outline of an example implementation of the TRACEalgorithm. In the following, a mathematical representation of the concept is provided. Thenotation is simplified by dropping the patient’s index. The focus is on the relationshipbetween features and concepts. Let ^^ ∈ represent the feature matrix,where ^^^,^indicates the f-th feature on the j-th time point in the sequence ending attime point t. Specifically, ^ ^^,^ is an element of the feature matrix where ^ is the window size, representing the number of preceding time points thatthe model considers to predict the event at the current time point, ^. contains features from the previous w days up to, but not including, day t.Let ^ ∈ × ^ represent the concept-feature mapping matrix, where:M^,^ = ^1 if feature f is part of concept c 0otherwiseThe aim is to compute the concept tensor ^^∈ℝ^×^, where K denotes the totalnumber of concepts. The transformation from the feature tensor ^^to the concept tensor ^^is given by: whererepresents the concept matrix for the time-series sequence ending at timet, and ^^^,^represents the concept vector at the j-th time point within thematrix ^^^ . M and K are predefined based on expert knowledge, and they can alsobe inferred through clustering algorithms. The table shown in Fig. 7 outlines theexample feature list used in the example implementation and their corresponding concepts.The encoder function, f, in the TRACE model encodes each concept at every timestep, and Z is computed by concatenating the representations across all conceptsand time steps within window w: where ^ denotes the concatenation operator for encoded representations of theconcepts at different time points.The decoder function, g, decodes the encoded representation Z to reconstruct anapproximation of the input data: ^^ = ^(^).In the example implementation, LSTM networks are used for both encoder anddecoder functions f^and g. Other models, e.g., CNN, LSTM-CNN, etc. can be used.Additionally, the Adam optimizer was leveraged during training due to its adaptivelearning rate properties, while other optimizers can be used as well.After training TRACE to generate representations (embeddings) for each concept, the trained encoders are saved. These encoders are then utilized for representation learning, where the autoencoder is trained to generate meaningful representations capturing important features and temporal dependencies from the input sequences.These learned representations (i.e., Z) can be used for downstream tasks such asforecasting or classification. In summary, after training TRACE to generate representations for each concept, themethod or system saves the trained encoders. These encoders can be then utilizedfor representation learning, wherein the autoencoder is trained to generate meaningful representations capturing important features and temporal dependencies from the input sequences. These learned representations can be leveraged for downstream tasks such as forecasting or classification using. To forecast the occurrence of future migraine attacks, a simpler model, such asRandom Forest, may be used on the temporal encoded feature concepts.In addition to forecasting migraine, the proposed system can be leveraged inpredicting other various health conditions (e.g., recurring medical symptoms) suchas periodic illnesses, also known as recurrent or episodic illnesses, are conditions that occur in cycles or with regular flare-ups and symptom-free periods. Examples of periodic illnesses include autoimmune diseases like lupus and Crohn’s disease,and other illnesses like Irritable Bowel Syndrome (IBS). These health conditionsalternate between phases of active symptoms (flare-ups) and remission, where symptoms decrease or temporarily disappear. The timing and triggers for these cycles vary, and due to their periodic nature, they are often managed with lifestyle modifications, dietary adjustments, and stress management, with additionaltreatments as needed during flare-ups. By applying the predictive system (TRACE),the timing of these flare-ups can be predicted, their triggers can be identified, andpatients can be helped to enhance their quality of life. Further, the predictive systemcan be applied to a sleep diary project that aims to improve sleep quality (as anexample of body behaviors) The proposed TRACE algorithm operates on data being used to train the autoencoder and, eventually, by processing corresponding input data. Therefore, longitudinal data (e.g., the input data / samples of input data) is collectedelectronically, e.g., from migraine patients, in both a structured (tabular) format andas user-provided text notes or doctor’s notes in electronic headache diaries. In this context, longitudinal data, also known as panel data, is a type of data that involves observations of the same subjects (such as individuals, groups, or organizations) repeatedly over a period of time. This kind of data collection allows researchers toanalyze changes over time and to understand dynamics across various subjects orentities. The idea is to observe how certain variables evolve, how interventions impact outcomes, or to identify trends that only appear over the course of severalmeasurements. Key characteristics and uses of longitudinal data include use of atime dimension, which allows tracking of changes in variables with respect to time, and repeated observation (where each subject in the dataset is observed multipletimes, enabling the analysis of intra-individual changes and variations. A time-seriesdataset may then be prepared using a sliding window technique.An output of the proposed algorithm is a patient-specific prediction, e.g., of future migraine attacks. The output may optionally further include an explanation, in whicha set of potential patient-specific migraine trigger factors can be derived usingmodel-agnostic XAI approaches such as SHAP or LIME. The output may optionallyfurther include a trigger identification (e.g., an explanation), which can be obtainedby analyzing temporal patterns, such as the specific day or time point when each trigger occurs. For example, fluctuations in caffeine intake between workdays andweekends may increase the likelihood of a migraine attack. For this purpose, themethod of Fig.1a may comprises providing 170, based on the set of Shapley values,information on an impact of features of the features related to the occurrence of the recurring medical symptom or body behavior on the prediction, and the method of Fig.2a may comprise computing 280, using the encoder machine learning modeland the prediction machine learning model, the set of Shapley values.The proposed concept includes two components – a training component, discussedin connection with Figs. 2a, 2b and 4, in which TRACE is trained, and anapplication / integration component, discussed in connection with Figs. 1a, 1b and 5,in which the trained encoders are integrated with a classifier, e.g., for forecasting migraine attacks. The training component, which is illustrated in Fig.4, comprisesone or more of the following operations and / or concepts. For example, the trainingcomponent may comprise grouping features into high-level concepts, in which thefeatures are organized into relevant, high-level concepts (i.e., categories), based oneither domain expert or using an existing clustering algorithm. The trainingcomponent may comprise splitting the concepts by time-step, thereby dividing each concept according to different time steps. For illustration, in Fig.4, three time stepsare shown, denoted as ^^, ^^. The training component may further compriseencoding time-step concepts, in which the encoder processes the time-stepconcepts and converts them into lower-dimensional representations. Therepresentations of individual time-steps (i.e., the single time-step embeddings) maybe concatenated to generate ^^^^^^^^ (the multi time-step embeddings), a lower-dimensional representation for each concept. The training component may furthercomprise decoding the lower-dimensional representation to reconstruct the originaldata. Using the latent space, the decoder may reconstruct an image of the originaldata by reversing the steps of the encoder. The integration / application componentmay comprise generating meaningful representations, in which the trained encodersare used to create meaningful representations (embeddings) that capture importantfeatures and temporal dependencies from the input concepts. The integration / application component may further comprise predicting future migraine attacks (or other recurring medical symptoms or behaviors), by using the learned representations ^^^^^^^^as inputs to a classifier, such as Random Forest, to predict the future recurring medical symptoms or behaviors, such as future migraineattacks. The integration / application component may optionally further compriseusing model-agnostic explainable AI (XAI) approaches to identify a set of potential patient-specific migraine triggers, with a focus on time-sensitive patterns. By analyzing time-specific variations, such as changes in routine or environmental factors, this approach uncovers triggers that are not only unique to each patient but may also fluctuate based on the timing of exposure, helping to better understand and anticipate migraine onset.Various examples of the proposed concept may include a temporal alignmentcomponent that ensures that the latent representations and their associatedconcepts remain consistent and relevant as the data evolves during the trainingphase. Various examples of the proposed concept may include a contextualanalysis module that assesses the relevance and relationships of features based on the data, application context, and expert knowledge. This module groups features into concepts and allows the integration of new concepts as new features areintroduced. Various examples of the proposed concept may perform dynamic groupadjustment, in which machine learning clustering techniques are used todynamically adjust the concept groups. For instance, if certain concepts become more relevant or change their relationships, the grouping algorithm updates the groupings accordingly. TRACE enhances encoding and decoding accuracy through temporal alignmentand contextual embeddings. The proposed method clusters features into conceptsto allow the autoencoder to generate relevant feature embeddings. Additionally, it preserves the temporal nature of the input data, ensuring that latent representations and their associated concepts remain consistent and relevant as the data evolves during training. This innovative approach allows for more accurate and contextually aware modeling of complex, time-dependent data. TRACE can handle a variety of features, e.g., headache-related features, lifestyle factors, medications, medical exam results (such as thyroid function tests, Blood pressure readings, blood sugar levels, etc.), features extracted from medicalimaging, and features collected from sensors. TRACE groups features into conceptsand then generates embeddings for these concepts, rather than for the raw features. This approach enhances the embeddings, enabling them to learn more usefulpatterns from relevant features. Previous research, on the other hand, focuses onlimited set of features and use conventional machine learning models to forecast migraine attacks. Existing concept bottleneck (CBM) models in the literature first infer an intermediateset of human-defined concepts, which are then used to predict the final output in aninteractive manner. These models have been applied to image data, such as kneex-ray images (see Koh, Prang Wei, et al. "Concept bottleneck models.") and medicaltime-series data where features are grouped into concepts that correspond tomeaningful clinical ideas (see Wu, Carissa, et al. "Learning optimal summaries ofclinical time-series with concept bottleneck models."), CBMs demonstrated improved performance and explainability. Further, CMBs allow users to directlymodify concept predictions and have those changes propagated to the finalprediction. This capability enables real-time corrections by experts, e.g., a radiologist can correct mistakes in concept predictions, thereby enhancing user interaction and accuracy. However, such models lack the concept embedding and the temporal alignment components thus they can be less efficient (in terms of speed and accuracy) than TRACE (see Koh, Prang Wei, et al. "Concept bottleneckmodels" and Wu, Carissa, et al. "Learning optimal summaries of clinical time-serieswith concept bottleneck models.") The model generated using TRACE was benchmarked against LSTM, CNN, a hybrid LSTM-CNN model, and a stand-alone Random Forest. The evaluations are conducted on two datasets: the 2-day forecast and the 3-day forecast. The datasets were split based on user ID, ensuring that the training, validation, and test sets consisted of mutually exclusive users. The data was divided into 70%^for training and 30%^for validation and testing. Training and testing predictive models on different user sets is considered a challenging task, yet it is necessary for robust real-world applications. Ideally, it isbeneficial to leverage user’s historical data to forecast their future headacheepisodes. However, when a user is new to the system or does not have enoughhistorical records, then data from existing users can be leveraged to enhance migraine predictions. This strategy not only tests the model’s capacity to generalize across a variety of users but also strengthens its practicality, especially in real-world scenarios where new users are consistently joining the system or existing users lack sufficient historical records.The performance of each method is evaluated over 5^random runs using precision,recall, macro F1 score, weighted precision, weighted recall, weighted F1 score, and per-class recall (also known as true positive rate for each class). The Macro F1 score calculates the F1 score for each class independently and then averages them, treating all classes equally regardless of size. Macro F1 score is sensitive to minority class performance, making it ideal for ensuring balanced performance across all classes, especially for imbalanced datasets.The TRACE model is benchmarked with several models previously leveraged forEHR time series forecasting. These include LSTM, CNN, a hybrid LSTM-CNNmodel, and a stand-alone Random Forest (see B. Shickel et al.: “Deep ehr: a surveyof recent advances in deep learning techniques for electronic health record (ehr)analysis” and Francois Chollet. Deep learning with Python.). Here, a high-leveloverview of each of these competing methods is provided.A first benchmark method uses an LSTM (Long Short-Term Memory). LSTM is atype of Recurrent Neural Network (RNN) that is capable of learning long-term dependencies in sequence data. This is useful for time series forecasting, where the temporal dependencies are crucial for making accurate predictions. The LSTM cell consists of three main components: the forget gate (ft), the input gate (it), and the output gate (ot). These gates collectively decide how much information to keep from the past, how much current information to let in, and how much information to output to the next time step (see P. L Benitez et al: “An experimental review on´ deeplearning architectures for time series forecasting”). The equations governing theLSTM operations are as follows: ^^ = ^^^^[ℎ^^^ + ^^] + ^^^^^ = ^(^^|ℎ^^^ + ^^| + ^^)^^ = ^(^^[ℎ^^^ + ^^] + ^^)^^ = tanh(^^[ℎ^^^ + ^^] + ^^)^^ = ^^ ⊙ ^^^^ + ^^ ⊙ ^^ℎ^ = ^^ ⊙ tanh(^^)Here, Xt is the input at time step t, ht−1 is the hidden state from the previous time step, and ct−1 is the cell state from the previous time step. The weights (W) andbiases (b) are the model parameters learned during training. The σ functionrepresents the sigmoid activation function, and the tanh function represents thehyperbolic tangent activation function. The operations are element-wise. The model trained using the Adam optimizer and is compiled with a categorical cross-entropy loss function: Nis the number of samples in the dataset.• M is the number of classes in the classification task.• yij is the true label of sample i for class j. It is 1 if the sample belongs to classj and 0 otherwise.• y0ij is the predicted probability that sample i belongs to class j according tothe model. In the context of time series forecasting, the LSTM model is trained on a sequence of historical data, and once trained, it can predict future values. The ability of LSTM to remember long-term dependencies makes it a popular choice for such tasks. A second benchmark method uses an CNN (Convolutional Neural Network).Additionally, a CNN architecture is employed as a benchmark model for the time-series forecasting task. The model begins with an input layer that takes a sequence of data with a specific shape (batch size,sequence length,in This is followed by convolutional layers that extract features from the input data using a convolution operation (see I. Koprinska et al.: “Convolutional neural networks for time series forecasting“): where Xt represents the input at time step t, W is the filter (or kernel), b is the bias, fis the ReLU activation function: f(x) = max(0,x), and k is the size of the filter (orkernel). To reduce computational complexity, pooling layers are used to downsample the feature maps generated by the convolutional layers, often using max pooling. The output of these layers is then flattened into a vector that can be fed into fully connected (FC) layers. These layers learn complex patterns in the features extracted by the convolutional layers, using a ReLU activation function: where Xflattenrepresents the flattened feature maps, WFCare the weights, bFCare thebiases, and f is the ReLU activation function.The final predictions are produced by the output layer, which uses a softmax activation function: where Ypredare the predicted probabilities for each class, Woutare the output layer weights, and boutare the output layer biases. Similar to LSTM, this model is compiled with a categorical cross-entropy loss function and the Adam optimizer.A third benchmark method uses hybrid model including a LSTM and a CNN. Theabove two architectures are combined to create a hybrid LSTM-CNN model forbenchmarking. Combining LSTM networks with CNN has emerged as a powerful approach for time series forecasting. LSTMs are well-suited for capturing long-term dependencies in sequential data, while CNNs excel at capturing local patterns. By combining these two architectures, the model can effectively capture both short- term and long-term dependencies in the time series data. Where, the LSTM layer is first used to learn temporal dependencies from the input data, and the output of the LSTM layer is fed into the CNN layers to capture spatial patterns. This combined architecture allows for more robust feature extraction and has been shown tooutperform traditional models in various time series forecasting tasks (see C. Lin etal.: “Early diagnosis and prediction of sepsis shock by combining static and dynamic information using convolutional-lstm”, K. Li et al.: “Convolutional recurrent neuralnetworks for glucose prediction” and G. Harerimana et al.: “Hsga: A hybrid lstm-cnnself-guided attention to predict the future diagnosis from discharge narratives”).A third benchmark method uses a random forest. Random forest is a powerfulmachine learning algorithm widely used for time series forecasting in various domains, including Electronic Health Records (EHR). By leveraging the ensemble of decision trees, random forest models can capture complex temporal patterns and relationships within the patient data, providing accurate forecasts for healthcareproviders to make informed decisions (see S. Wongvibulsin et al.: “Clinical riskprediction with random forests for survival, longitudinal, and multivariate (rf-slam) data analysis”, J. Wang et al.: “. Risk assessment of coronary heart disease based on cloud-random forest” and M. J. Kane et al.: “Comparison of arima and random forest time series models for prediction of avian influenza h5n1 outbreaks”). LSTM, LSTM-CNN, and TRACE models were implemented using Python version 3.11.5, Tensorflow version 2.15.0, and a GeForce RTX 2080^GPU model. Different sets of hyper-parameters (e.g., embedding dimension, convolution kernel sizes, LSTM hidden units, batch size, etc.) were explored using random search. For eachof these models, the maximum epoch number is set to 50. Finally, the following earlystopping strategy was implemented during training: if the validation recall did not improve over 10^epochs (i.e.,patience parameter), the training process was halted. All competing models leveraged weighted training to address the issue of class imbalance.Figs. 8a and 8b compare the per-class recall across five distinct runs for competingmethods using 2-day forecast and 3-day forecast datasets, each initiated with aunique random seed. In these datasets, the models leverage relevant migrainefeatures from the preceding 2 or 3 days, to predict migraine attacks the following day. The average migraine recall rates for the 2-day forecast dataset are as follows: 0.22 for Random Forest, 0.54 for CNN, 0.58 for LSTM, 0.60 for LSTM-CNN, and 0.67 for TRACE. For the 3-day forecast dataset, the average values are: 0.19 for Random Forest, 0.62 for CNN, 0.58 for LSTM, 0.54 for LSTM-CNN, and 0.68 for TRACE. In both datasets, the TRACE achieves the highest migraine recall of almost 70%, indicating that nearly 70% of all relevant migraine attacks were correctlyforecasted. However, it is crucial to note the trade-off between forecasting migrainesand no headache days. For example, the Random Forest model demonstrates the highest recall for headache-free days but the lowest recall for migraine episodes.Migraine recall was prioritized due to the potentially higher costs associated withunforeseen migraine attacks. For example, migraine sufferers can mitigate the severity of an attack by adhering to specific routines, such as ensuring adequate sleep, following a particular diet, and avoiding excessive alcohol consumption, which is known to exacerbate headache frequency and intensity. Figs. 9a, 9b and 9c present a comparative analysis of three key performance metrics: precision, recall, and macro F1 score. Figs. 9a and 9b show precision, recall, and macro F1 score values across five different runs for the competing models: Random Forest, CNN, LSTM, LSTM-CNN hybrid, and TRACE. Theevaluation is performed using two datasets: 2-day forecast and 3-day forecast. Fig.9c also shows precision, recall, and macro F1 score values across five different runs for the competing models: Random Forest, CNN, LSTM, LSTM-CNN hybrid, and TRACE. The evaluation is performed using two datasets: 2-day forecast and 3-day forecast.Fig. 10 shows the corresponding weighted key performance metrics: precision,recall, and F1 score values across five different runs for the competing models: Random Forest, CNN, LSTM, LSTM-CNN hybrid, and TRACE. The evaluation isperformed using two datasets: 2-day forecast and 3-day forecast. Similarly to Figs.9a, 9b and 9c, the performance metrics are computed over 5 distinct runs on 2-day and 3day forecast datasets. All competing methods demonstrate comparable precision and F1 scores, with TRACE achieving the highest recall rates for both datasets.One metric being used is the macro-averaged F1 score. When calculating themacro-averaged F1 score across multiple classes, the F1 score is computed foreach class c, then average these scores. It is defined as F1_macro Where N is the total number of classes and ^1^ is the F1 score for class c.In scikit-learn, ‘f1 score‘ with ‘average=’macro’‘ implements this macro-averaging process automatically. It computes F1 c for each class and then averages these scores to give you ‘f1 test macro‘, the macro-averaged F1 score across all classes.Using this metric ensures that all classes are treated equally, which is especiallyvaluable when assessing minority class performance. Another metric being used is the weighted precision score. The precision score for each class is computed and then averaged by the support, i.e., the number of true instances for each label. where ^^^^^^^^^^ is the precision of class i and ^^^^^^^^ is the number of trueinstances of class i. Another metric being used is the weighted recall score. The recall score for each class is computed and then averaged by the support, i.e., the number of true instances for each label. where ^^^^^^^is the recall of class i and ^^^^^^^^is the number of true instances of class i.Another metric being used is the weighted F1 score. The function f1score fromsklearn.metrics is used to compute the F1 score. For weighted F1 score, theparameter average=’weighted’ is added, which calculates the F1 score for each classindividually and then computes the weighted average of these scores based on thesupport (the number of true instances for each class), see scikit-learn contributors.scikit-learn: Machine Learning in Python, 2022 . where ^^^^^^^^^^ is the precision for class c, ^^^^^^^ is the recall for class c.The weighted average F1 score is as: where ^^^^^^^^ is the number of true instances for class c.Another metric being used is the per-class recall score. Considering “no headache” as the negative class and “migraine” as the positive class, the per-class recall iscomputed as follows. For the ”no headache” class (negative class): For the ”migraine” class: The table shown in Fig.11 summarizes the average training time in minutes for eachof the competing model and the standard deviation in minutes across 5 runs for each model. In the 2-day forecast dataset, the LSTM-CNN model had the longest training time with significant variance. Individual runs showed that the training time ranged from 67 to 230 minutes. However, this model exhibited less variance in training time for the 3-day forecast. It is assumed that this is because the 3-day forecast dataset includes more historical data, which may provide a more stable training process. The training time for TRACE includes the total duration needed to generate the concept embeddings and to train the random forest model on these embeddings. The model operates in parallel on each concept to create the corresponding embeddings. While training the random forest typically takes less than five minutes on average, generating the concept embeddings is a more time-consuming step. Consequently, TRACE generally requires more training time compared to othermodels. Lastly, Fig. 12 provides a global overview of feature importance derivedfrom the Random Forest model leveraged in the TRACE algorithm across 5 runs onthe 3-day forecast dataset.To understand the importance of the different concepts, an ablation study wasconducted to evaluate the performance of the TRACE method when each conceptis considered individually. The goal is to determine the impact of each featureconcept on the overall performance. Fig. 13 shows an ablation analysis illustratingthe recall per class for individual feature concepts and their combination using the3-day forecast dataset acrossfive different runs.Considering each concept, the health factor concept, denoted as Chealthfactor, achieves the highest recall for the no-headache class, while, the medication concept, denoted as Cmedication, provides the highest recall for the migraine class. While each concept individually demonstrates acceptable performance, theircombination results in an enhanced recall for the migraine class. Fig. 14 showsprecision, recall, macro F1 score, and the corresponding weighted values across five different runs for individual feature concepts and their combination using the 3- day forecast dataset. Some examples of the proposed concept provide personalized explanations usingSHAP. In healthcare, machine learning model interpretability and transparency are beneficial. Knowing why a model makes a specific prediction can directly impact patient care, regulatory compliance, and trust in AI-driven systems. Using SHAP for explainable AI (XAI) methods allows to pinpoint specific triggers at the patient level, supporting precision medicine initiatives. For instance, SHAP values can reveal individual sensitivities, such as a patient’s increased likelihood of migraines following caffeine consumption, enabling tailored recommendations based on personal triggers. One of the key challenges to tackle is model explainability to ensure safe model deployment and build user trust. Consequently, various eXplanation Artificial Intelligence (XAI) techniques have been proposed to gain insights and provide local and global explanations for models behavior. In Duell et al.: “A Comparison of Explanations Given by Explainable Artificial Intelligence Methods on Analysing Electronic Health Records”, the researchers evaluated three XAI techniques for analyzing EHRs: SHapley Additive exPlanations (SHAP) (see also Lundberg et al.: “A Unified Approach to Interpreting Model Predictions”), Local Interpretable Model- agnostic Explanations (LIME) (Ribeiro et al.: “Why Should I Trust You. Explaining the Predictions of Any Classifier), and Scoped Rules (see Ribeiro et al.: “Anchors: High-Precision Model-Agnostic Explanations). These techniques are featureattribution methods that assign weights to the features used for model predictions.Their benchmark focused on the ability of these methods to provide both local andglobal explanations, supporting human reasoning and improving trust in AI models.According to their analysis, SHAP was considered the most effective XAI method, offering clarity in communicating both local and global explanations. Further, SHAP visualizations effectively supported clinicians’ needs by providing reasoning and enhancing trust in human-AI collaboration.In Wu et al.: “Learning Optimal Summaries of Clinical Time-series with ConceptBottleneck Models, a different XAI approach was introduced. The researchers aimed to facilitate human inspection of clinical time-series prediction models by learning ‘concepts’ that group features into high-level categories. They proposed an optimization method to select the most important features within each concept, resulting in the learning of sparse prediction models, i.e., models that have a limited number of features. These models strike a balance between complexity and interpretability, allowing them to be effectively analyzed and understood by human experts, which is crucial in clinical settings where decisions based on machine learning need to be justified and understood by domain experts.In the following, the LSTM model is leveraged to generate SHAP per-user insightsof migraine trigger factors. SHAP calculates feature importance scores by considering the contribution of each feature to the model’s output. It assigns each feature an importance score by computing the average marginal contribution of that feature across all possible combination of features. It ensures a consistent and interpretable measure of feature impact based on game theory’s Shapley values.Figs. 15a to 15j display SHAP summary plots of the top 10 features influencingmigraine occurrences for 10 users who consistently used the app over an extended period. In Fig. 15a, the SHAP analysis indicates that higher medication usage is associated with a reduced likelihood of migraine occurrence. Furthermore, reduced sleep duration and quality increase the chances of a migraine attack for this user. In Figure 15b, the SHAP analysis reveals that caffeine overconsumption on days -2 and -3, coupled with reduced caffeine intake on day -1, is linked to a higher likelihood of experiencing migraines for this user, similar pattern is observed in other patients, see Figs.15c, 15e, 15f, 15g and 15j.Figs. 15a to 15j show SHAP summary plots illustrating the top 10 features and theirimpact on predicting migraine occurrences among users who consistently used theapp over an extended period. In thisfigure, (day -i) refers to the i’th day precedinga migraine attack. The user shown in Fig.15a actively engaged with the M-sense mobile app from April 13, 2017, to October 08, 2022, contributing a total of 1871 daily records. Analysis of SHAP top features revealed that medication use ranked prominently among the most influential factors in preventing migraine occurrences. Increased stress, alcohol consumption, and reduced sleep duration and quality on the day precedingthe attack contribute to a higher likelihood of migraine occurrence for this user. Theuser shown in Fig.15b actively engaged with the M-sense mobile app from January 27, 2018, to October 02, 2022, contributing a total of 1687 daily records. An analysis of SHAP top feature lists revealed that variations in normalized caffeine consumption are among the most influential factors in the occurrence of migraine attacks for this user. Additionally, temperature changes in the two days preceding a migraine attack significantly impact its onset. Furthermore, a decline in normalizedsleep duration on the day immediately prior to the attack indicates a higher likelihoodof experiencing a migraine. The user shown in Fig. 15c actively engaged with the M-sense mobile app from June 25, 2018, to March 02, 2022, contributing a total of 1287 daily records. The top features contributing to the likelihood of migraine occurrence include caffeine,alcohol, stress, and period (i.e., menstrual cycle). The user shown in Fig. 15dactively engaged with the M-sense mobile app from January 27, 2018, to September 13, 2022, contributing a total of 1529 daily records. The top features contributing to the likelihood of migraine occurrence include medication, caffeine, temperature, andstress. The user shown in Fig. 15e actively engaged with the M-sense mobile appfrom November 08, 2018, to October 05, 2022, contributing a total of 1294 daily records. The top features contributing to the likelihood of migraine occurrenceinclude stress, caffeine, medication, and alcohol consumption. The user shown inFig.15f actively engaged with the M-sense mobile app from January 11, 2019, to August 11, 2022, contributing a total of 1179 daily records. The top features contributing to the likelihood of migraine occurrence include stress, caffeine, and sleep (both quality and duration).The user shown in Fig. 15g actively engaged with the M-sense mobile app fromNovember 11, 2018, to September 30, 2021, contributing a total of 978 daily records. The top features contributing to the likelihood of migraine occurrenceinclude stress, caffeine, alcohol, and temperature. The user shown in Fig. 15hactively engaged with the M-sensemobile app from April 30, 2018, to October 09, 2022, contributing a total of 1394 daily records. The top features contributing to the likelihood of migraine occurrence include smoking, caffeine, temperature, andmedication. The user shown in Fig. 15i actively engaged with the M-sensemobileapp from July 18, 2017, to October 09, 2022, contributing a total of 1815 daily records. The top features contributing to the likelihood of migraine occurrence include stress, caffeine, medication, and sleep duration. The user shown in Fig.15j actively engaged with the M-sensemobile app from January 28, 2018, to September 11, 2021, contributing a total of 1297 daily records. The top features contributing to the likelihood of migraine occurrence include alcohol, caffeine, sleep duration, and stress.The observations indicate that higher temperatures on the day before a migraineincrease the likelihood of its occurrence. In contrast, lower temperatures two days prior seem to have a similar effect, see Figs.15a, 15b, 15d and 15f. Similar findings were reported in previous research, which investigated how both temperature levelsand fluctuations influence the probability of a migraine attack for some patients.Fig. 16 shows another illustration of an analysis of feature importance for a specificuser. The example shown in Fig. 16 shows that reduced caffeine consumption oneday before a migraine attack (day -1) is associated with a higher likelihood ofmigraine, while reducing caffeine two days prior (day -2) appears to be associated with no migraine. This suggests that variations in caffeine intake may contribute to migraine onset. Additionally, increased stress levels are observed to heighten the likelihood of migraine occurrence for this user.Stress was the most prevalent trigger for migraines in the dataset, with elevatedstress levels significantly increasing the likelihood of a migraine the following day, consistent with prior research. Sleep duration and / or quality emerged as anothercommon trigger in the dataset, with shorter sleep periods and lower sleep qualitysignificantly increasing the likelihood of a migraine the next day. Furthermore, the analysis reveals that, although common migraine triggers exist, individual triggerscan vary significantly, for example, patient-specific triggers were observed, such asperiod (i.e., menstruation) in Fig. 15c, smoking in Fig. 15h, and excessive alcoholconsumption in Fig.15j.The study demonstrates the effectiveness of TRACE in forecasting migraineattacks, achieving approximately 70% accuracy in 2-day and 3-day forecast datasets. This predictive capability significantly enhances the potential for proactive migraine management, which can greatly improve the quality of life for sufferers. By leveraging a comprehensive dataset and employing both global and local featureimportance analyses, critical predictive factors were identified and the variability ofmigraine triggers for different patients was underscored.TRACE may be refined to further improve its performance and adaptability across diverse patient datasets, such as including more diverse demographics, sensor data, and clinical variables such as brain imaging. A key direction involves conducting an in-depth analysis of complex feature engineering techniques to better understand the correlations and causations between features and migraine onset. Additionally, integrating real-time data collection and feedback mechanisms could enhance the model’s responsiveness and accuracy, ultimately contributing to more effective and tailored treatment strategies. The present invention may be implemented as a computer-implemented method, computer system (comprising one or more processors and one or more storage devices) configured to perform the computer-implemented method and / or as a computer program for performing the computer-implemented method. For example, the computer-implemented method may include one or more steps and / or operations discussed above. Various aspects of the present invention relates to machine learning. In particular, the model mentioned above may be a machine learning model. Machine learning is a branch of artificial intelligence that involves the development of algorithms and models that allow computers to learn and make predictions or decisions without being explicitly programmed. It focuses on creating systems that can improve their performance over time by learning from data. Training a machine-learning model refers to the process of teaching the model to make accurate predictions or decisions. During training, the model is exposed to a large amount of data, which is used to adjust the model's internal parameters or weights. The model learns patterns, relationships, or rules from the training data, allowing it to generalize and make predictions on new, unseen data. Training data is the set of examples or instances that is used to teach a machine-learning model. It is often labeled data, meaning that each example is associatedwith a known outcome or target value. The training data consists of both input features and the corresponding output or target variable. The model learns from this data by analyzing the patterns and relationships between the input features and the target variable. Training algorithms, such as supervised learning, semi-supervised learning, unsupervised learning or reinforcement learning may be used for training the machine-learning model. Machine-learning models, such as the machine-learning model being trained in the present disclosure, are often implemented as Artificial Neural Networks (ANNs), and in particular Deep Neural Networks, Support Vector Machines, Decision Tree models, or Random Forest models. Examples may involve or relate to computer programs, including program codes to execute one or more of the mentioned methods when the program is executed on a computer, processor, or other programmable hardware component. As a result,steps, operations, or processes from various methods described above can also beexecuted by computers, processors, or other programmable hardware components. Examples may additionally cover program storage devices, such as digital data storage media, which are machine-, processor-, or computer-readable and encode and / or contain machine-executable, processor-executable, or computer-executable programs and instructions. These devices may include or be digital storage devices,magnetic storage media like magnetic disks and tapes, hard disk drives, or opticallyreadable digital data storage media, for instance. Other examples encompass computers, processors, control units, field programmable logic arrays (FPLAs), field programmable gate arrays (FPGAs), graphics processing units (GPUs), application- specific integrated circuits (ASICs), integrated circuits (ICs), or system-on-a-chip (SoC) systems that are programmed to carry out the steps of the aforementioned methods. In simpler terms, examples may involve computer programs and storage media comprising computer programs, as well as hardware components like processors and control units, which can be programmed to execute the methods described above. When certain aspects are mentioned in relation to a device or system, they should also be considered as descriptions of the corresponding methods. For example, a block, component, or functional aspect of the device or system may correspond to a method step or feature of the related method. Therefore, aspects described regarding a method should also be understood as depicting a corresponding element, property, or functional feature of the corresponding device or system. In simpler terms, if something is described in relation to a device or system, it can also be applied to the corresponding method, and vice versa.Many modifications and other embodiments of the invention set forth herein willcome to mind to the one skilled in the art to which the invention pertains having thebenefit of the teachings presented in the foregoing description and the associateddrawings. Therefore, it is to be understood that the invention is not to be limited tothe specific embodiments disclosed and that modifications and other embodimentsare intended to be included within the scope of the appended claims. Althoughspecific terms are employed herein, they are used in a generic and descriptivesense only and not for purposes of limitation.

[0002] L i s t o f r e f e r e n c e s i g n s10, 20 System12, 22 Interface circuitry14, 24 Processor circuitry16, 26 Memory / storage circuitry110 Obtaining input data120 Dividing features into groups130 Encoding groups of features intoembeddings 140 Concatenating embeddings150 Inputting embeddings intoprediction machine learning model 160 Providing a prediction170 Providing an explanation210 Obtaining training data withsamples of input data 220 Determining a mapping betweenfeatures and groups of features 230 Dividing the features of thesamples of input data into the groups of features 240 Training an autoencoder(including an encoder ML model) 250 Concatenating embeddings260 Training a prediction ML model270 Providing the ML model(s)280 Computing / providing Shapleyvalues

Claims

C l a i m s1. A computer-implemented method for optimizing a prediction of an occurrenceof a recurring medical symptom or body behavior of a human being usingmachine learning, the method comprising: obtaining (110) input data comprising features related to the occurrence of the recurring medical symptom or body behavior;dividing (120) the features included in the input data into two or more groupsof features; encoding (130), using a trained encoder machine learning model, thefeatures of the two or more groups of features into two or more embeddings, with each embedding representing a group of features; inputting (150) the two or more embeddings into a prediction machinelearning model being trained to predict the occurrence of a recurring medical symptom or body behavior based on the two or more embeddings; andproviding (160) a prediction of the recurring medical symptom or bodybehavior based on an output of the prediction machine learning model.

2. The method according to claim 1, wherein the input data is longitudinal inputdata comprising a plurality of observations of the same subject over a plurality of points in time, with the two or more embeddings being multi time-stepembeddings each representing a group of features over multiple points intime.

3. The method according to claim 2, wherein the two or more multi time-stepembeddings each comprise a concatenation of single time-step embeddingsoutput by the trained encoder machine learning model, with each single time-step embedding representing a point of time.

4. The method according to one of the claims 2 or 3, wherein the two or moremulti time-step embeddings are encoded based on a window of points of timewithin the plurality of points in time.

5. The method according to one of the claims 1 to 4, further comprisingconcatenating (140) the two or more embeddings to obtain a concatenatedembedding representing the groups of features, and providing theconcatenated embedding as input for the prediction machine learning model.

6. The method according to one of the claims 1 to 5, wherein the prediction ofthe recurring medical symptom or body behavior is a prediction of a migraine attack, with the recurring medical symptom being the migraine attack.

7. The method according to one of the claims 1 to 6, wherein the predictionmachine learning model is implemented as a random forest.

8. The method according to one of the claims 1 to 7, wherein the two or moregroups of features comprise at least one group comprising topically relatedfeatures, and / or wherein the method comprises providing (170), based on a set ofShapley values, information on an impact of features of the features relatedto the occurrence of the recurring medical symptom or body behavior on theprediction.

9. A computer-implemented method for training at least one machine learningmodel for use in predicting occurrence of a recurring medical symptom or body behavior, the method comprising: obtaining (210) a set of training data, the set of training data comprising aplurality of samples of input data, with each sample of input data comprising features related to the occurrence of the recurring medical symptom or bodybehavior;dividing (230) the features included in the samples of input data into two ormore groups of features; training (240) an autoencoder comprising an encoder machine learning model and a decoder machine learning model using the set of training data,wherein the autoencoder is trained to encode, using the encoder machine learning model, the respective groups of features into an embedding representing the respective group of features, with the resulting embeddings having a lower dimensionality than the respective groups of features represented by the embeddings, and to decode, using the decoder machine learning model, the embeddings to restore the groups of features; and providing (270) the encoder machine learning model of the autoencoder foruse in encoding the features of the two or more groups of features into embeddings.

10. The method according to claim 9, wherein the autoencoder is trained usingunsupervised learning, and / or wherein the encoder machine learning model is at least partially trained based on input data samples related to a target human being, and / orwherein the encoder machine learning model is at least partially trained based on input data samples related to a plurality of human beings.

11. The method according to one of the claims 9 or 10, wherein the features aredivided into the two or more groups of features according to a mapping between features and groups of features, wherein the method comprises determining (220) the mapping using aclustering algorithm, or wherein the mapping between features and groups of features is pre-defined according to expert knowledge.

12. The method according to one of the claims 9 to 11, wherein the methodcomprises training (260) a prediction machine learning model to predict the occurrence of the recurring medical symptom or body behavior based on theembeddings output by the encoder machine learning model, and providing (270) the prediction machine learning model together with the encodermachine learning model.

13. The method according to one of the claims 9 to 12, wherein the methodcomprises computing (280), using the encoder machine learning model andthe prediction machine learning model, a set of Shapley values, the set ofShapley values representing an impact of features of the features related to the occurrence of the recurring medical symptom or body behavior on aprediction provided by the prediction machine learning model.

14. A computer program comprising instructions which, when the program isexecuted by a computer, cause the computer to carry out at least one of the methods according to one of the claims 1 to 13.

15. A system (10; 20) comprising interface circuitry (12; 22), machine-readableinstructions, and processor circuitry (14; 24) to execute the machine-readableinstructions to carry out at least one of the methods according to one of theclaims 1 to 13.

Images