Intelligent Clinical Information Wizard (ICIW)

PhD project information


This PhD work will develop a new multimedia environment for intelligent access to clinical patient information, through the integrated application of a set of advanced AI and visualization technologies.

This highly usable interactive environment will allow doctors to easily access clinical information about their patients, overcoming the serious information overload problem they are currently facing when accessing Patient Health Records.

The PhD thesis will apply information extraction to derive significant clinical information from textual sources, advanced visualization and infographics to synthesize pathology and specialty based patient information, Internet bot and Dialogue Natural Language Processing(NLP) based technology for easy interactive access to clinical information.


  • Industrial partner: Exprivia SpA
  • Academic/research partner: University of Trento
  • Number of available PhD positions: 1
  • Duration: 3 years
  • Innovation Focus Areas(s): Digital Wellbeing
  • DTC location: Trento
  • This PhD will be funded by EIT Digital and Exprivia SpA

PhD thesis motivation and innovation valorisation


The significant increase in the use of digital solutions in healthcare allowed in the past few years the development of large Patient Health Records at Hospital, Regional and National level, collecting clinical evidence produced in the process of care.

Unfortunately, a large portion of this information is still represented by documents written in natural language, with a serious lack of structured, machine actionable, information. In addition, the amount of documents stored in these repositories is significant, reaching many tens of documents per year for the most critical patients, such as chronic and/or multimorbidity patients. As a result, doctors are suffering a typical information overloading condition. In order to build a complete and reliable idea of the patient clinical condition and history, a significant number of documents are available, but doctors don’t have the time to read and analyse all this information.

The objective of this PhD thesis is to exploit the enormous amounts of important, sometimes vital, clinical information for patient care by developing an advanced intelligent environment able to support medical doctors in accurate, fast and reliable access to the clinical information required in their daily practice.


The PhD thesis will develop a number of different solutions, to be integrated in a usable visual and voice-enabled environment. The key problems to be solved are the following:

  • Extracting relevant medical information from natural language textual sources
  • Building access tools to extract patient information from standard repositories and non-standard data bases
  • Building smart visual representations in order to provide synthetic information on patient history and their clinical data. Specialised visual representations will be elaborated, related to specific pathologies and specialties, using a combination of infographics and anatomical maps
  • Integrating voice recognition and dialogue, NLP -based technology to support automatic question answering to access patient specific information (ex. “please plot past glucose test results for this patient”, “what’s the last value for ejection fraction?”)

Expected academic outcomes

From a scientific perspective, the major challenges regard the definition of methods for accessing the desired information. One powerful solution is based on automated question answering (QA). The IBM Watson system [Ferrucci et al., 2010] has shown the effectiveness of using such approach in the Jeopardy! Challenge. Since then, simpler but more advanced QA methods based on Neural Models have been developed, e.g., [Yih et al., 2014, Severyn and Moschitti, 2015, Severyn and Moschitti, 2016]. An even more advanced solution to find information is the used of QA-based Dialog systems [Yao at al., 2015]. These also can be based on Neural Networks, e.g., using sequence to sequence neural models [Sutskever et al., 2014], extended with dialog manager components.

The academic outcomes will include papers publications - at least three papers are envisaged - and participation in relevant academic events, in particular NLP and healthcare related conferences.

Concrete innovations expected as the outcome of the proposal

The ICIW system is intended to support medical professionals in acquiring a synthetic view of the patient clinical history, through domain/pathology specific visualizations (extracting and selecting the information needed for a specific discipline and/or pathology) and dialogue based human-computer interaction. In particular, we expect this system to be extremely useful for family doctors acquiring new patients and for specialized consultants providing second opinion: today these professionals are supposed to read tens of PDF documents stored in EHRs (Electronic Health Records) in order to understand  the patients condition and history (they have no time for this, so they are forced to work with limited and uncertain information provided by patients or relatives).

New Products: Exprivia plans to industrialize and propose this Intelligent Clinical Information Wizard as an add on to any existing Electronic Medical record and/or Electronic Health Record, as a tool able to automatically synthesize, represent and personalize the relevant clinical patient information. The way this will integrate into existing systems is very simple: from the backend, the ICIW will index the clinical document sources available at hospital/territory level in order to build a structured history of the patient, while from any existing application context (from EMR or HER GUIs) a simple parametric URL invocation will activate the ICIW synthetic patient history visualization and the dialogue interface;
Improved Products: the Intelligent Clinical Information Wizard will be integrated his Electronic Health Record (EHR) module. The Exprivia EHR is currently composed by eRepository (a IHE compliant XDS Repository/Registry), eArchive (a Dicom compliant image/video/waveform archive), ePrivacy (a GDPR compliant consent and sensible information access manager) and eViewer (an HTML5 responsive document/image/video/waveform visualizer). The ICIW will be configured as an innovative additional module of the EHR: the information sources will be eRepository and eArchive, the user functionality will be an extension of eViewer, while the privacy constraints will be managed by integrating the ePrivacy module.

A first version of the Clinical Information Wizard, including the Information extraction and Visualization technology, is expected to be available within the first two years of activity, while the most advanced Q&A and Dialogue based functionality is expected to be industrialized immediately after the PhD conclusion.

Expected impact of the PhD outcomes with respect to their business line

The key objective of the ICIW is to provide clinicians an easy, time saving and effective way to acquire significant information extracted from large databases of clinical documents. These technologies are intended to overcome a well known information overload problem arising in the interaction with poorly structured clinical repositories, such as hospital, regional and national Electronic Health Records (EHRs, in Italian terminology FSE -Fascicoli Sanitari Elettronici). Although hospital repositories are included in the clinical information systems in hospitals, the EHR market is dominated by territory healthcare management organizations, mainly at the level of regions and nations, the EHR being a large archive of documents and images/waveforms storing the entire clinical history of a patient.  Exprivia is active in this market as provider of software and services for around 15 hospitals, 4 Italian regions and a large Mexican healthcare public insurance serving 13 millions citizens (ISSSTE - Instituto de Seguro Social de Trabhadores del Estado).

An important impact is expected on multiple levels:

  • the envisaged tool will be proposed to all existing customers (more than 200 hospitals, 5 main regions and one national institution using Exprivia healthcare products)
  • Product line: the tool will greatly increase the Exprivia products competitiveness, thanks to a unique set of advanced, AI based features
  • Business/market: the envisaged tool will be proposed to all existing customers (more than 200 hospitals, 5 main regions and one national institution using Exprivia healthcare products);
  • as a standalone product, the ICIW will be proposed to all Italian Regions and to foreign nations/regions/hospitals (such as the Mexican ISSSTE) in order to enforce their EHRs
  • Professionals/patients: the new tool will provide clinicians a very innovative, useful and efficient medium to support easy access to patient information, therefore enhancing the overall quality of the medical care for the patients.

PhD thesis time-line and milestones

M6Architecture design and selection of methods and technologies for Information Extraction, Q&A Dialogue and Visual/Infographics representations. Study and selection of reference clinical standard terms resources (such as SNOMED, ICD9/ICD10, MESH, ACR-NEMA, etc.).
M12 First system prototype implementing the Information Extraction tool specialized in a medical subdomain (Cardiology) integrating the use of one or more standard medical terms resources.
M18Prototype of the Wizard implementing Information Extraction and Visualization technology. The prototype will demonstrate advanced visualization techniques accessing structured data extracted from a document repository containing patients clinical history.
M24Development of the first Q&A and dialogue module prototype, able to answer questions on patient’s clinical history by accessing structured data extracted from repository of clinical documents.
M30Second Wizard prototype integrating Information extraction, Advanced Visualization, Q&A and dialogue modules, able to provide the complete system functionality.
M36Final version of the Intelligent Clinical Wizard and report on the experimental use of the system in a real hospital environment by clinical users. The report  will include a Usability Assessment Report collecting feedback from the users on Effectiveness, Efficiency, Accuracy and Usability of the tool.

International mobility plan

The mobility destination(s) will be chosen with the objective of combining the presence of Exprivia branches (Germany, France, Spain, Netherlands, Belgium, Poland and UK) and academic centers collaborating with DISI.

Will the PhD Student do the Business Development Experience at Industrial Partner premises?



  • D. Ferrucci, E. Brown, J. Chu-Carroll, J. Fan, D. Gondek, A. Kalyanpur, A. Lally, J. W. Murdock, E. Nyberg, J. Prager, N. Schlaefer, C. Welty, "Building Watson: An Overview of the DeepQA Project", AI Mag., vol. 31, no. 3, pp. 59-79, Fall 2010.
  • Yih , W.-T., He, X., and Meek, C. Semantic parsing for single-relation question answering. In ACL, 2014.
  • Sutskever, I., Vinyals, O., and Le, Q. (2014). Sequence to sequence learning with neural networks. In Advances in Neural Information Processing Systems (NIPS 2014).
  • Yao, K., Zweig, G., and Peng, B. . Attention with intention for a neural network conversation model, 2015.
  • Severyn, A. and Moschitti , "Learning to rank short text pairs with convolutional deep neural networks.", in SIGR, 2015.
  • Severyn, A. and Moschitti, A.. Modeling relational information in question-answer pairs with convolutional neural networks, 2016.


If you are interested in applying to this position, please follow this two-step process: 

  1. Complete the EIT Digital application form here;
  2. Apply on the relevant University system at this link.

© 2010-2020 EIT Digital IVZW. All rights reserved. Legal notice