Research areas

  • Clinical decision support

    Clinical decision support

    Cli­nical deci­sion sup­port refers to the process of pro­vi­ding impor­tant infor­ma­tion to healthcare pro­fes­sio­nals (e.g., cli­nicians) and patients alike to help inform deci­sion-making in order to improve healthcare outco­mes and deli­ve­ry. In the con­text of healthcare infor­ma­tics, elect­ro­nic cli­nical deci­sion sup­port sys­tems (CDSS) will typically con­sist of a medical know­ledge base (inclu­ding patient infor­ma­tion), an infe­rence met­hod to fil­ter and 'learn' from the data, and a presentat...


  • Consumer health informatics

    Consumer health informatics

    Con­su­mer health infor­ma­tics is a speci­fic focus wit­hin bio­me­dical infor­ma­tics per­tai­ning to patients and their use of health-­re­la­ted infor­ma­tion. Con­su­mer health infor­ma­tics addres­ses a range of topics, inclu­ding health lite­racy, per­so­nal health records (PHRs), patient gene­ra­ted data, mHealth, human com­pu­ter inte­rac­tion (HCI), and con­su­mer educa­tion. Research wit­hin this area is hence concer­ned with inte­gra­ting patient infor­ma­tion needs and pre­fe­rences into effec­tive design of proces­ses an...


  • Knowledge discovery and data mining

    Knowledge discovery and data mining

    Know­ledge Disco­very and Data Mining is the process of auto­ma­ting com­pu­ta­tio­nal and sta­tis­tical ana­ly­sis of bio­me­dical data­sets (data mining) to derive use­ful insight in diag­no­sis, the­ra­py, and healthcare costs (know­ledge disco­ve­ry). It requi­res robust met­hods in such ana­ly­tical sta­ges as data­base mana­ge­ment, data pre-­proces­sing, fea­ture selec­tion, infe­rence and prior know­led­ge, model selec­tion, visua­liza­tion, and results ana­ly­sis. Mining and disco­very tech­niques can be applied in medicine ...


  • Knowledge representation and semantics

    Knowledge representation and semantics

    Know­ledge repre­sen­ta­tion is a field wit­hin AI that deve­lops mac­hi­ne-­rea­dable for­mats for infor­ma­tion. This mac­hi­ne-­rea­dable infor­ma­tion can then be used by mac­hi­nes eit­her to per­form tasks that were once per­formed only by humans or to assist humans in per­forming tasks more effec­ti­ve­ly. In bio­me­dical infor­ma­tics, know­ledge repre­sen­ta­tion is crucial for deve­lo­ping com­pu­ter-ai­ded detec­tion/­diag­no­sis (CAD) sys­tems. A chal­lenge in this field is to deve­lop repre­sen­ta­tions that are not only reada...


  • Medical imaging informatics

    Medical imaging informatics

    Image ana­ly­sis has become a power­ful tool to diag­no­se, eva­luate and fol­low up ove­rall disease chan­ge. As cli­nical needs evol­ve, quan­ti­ta­tive ima­ging tech­niques have been in con­ti­nuous deve­lop­ment to ana­lyze and repre­sent chan­ges on ima­ging fin­dings that can be used to bet­ter pro­duce insights about treat­ment effec­ti­ve­ness and unders­tand disease progression.

    Ima­ging tech­niques such as MRI, CT and PET com­bi­ned with modern com­pu­ter-­ba­sed image ana­ly­sis tech­niques (com­pu­ter vision, mach...


  • Natural language processing

    Natural language processing

    Natu­ral lan­guage proces­sing encom­pas­ses the auto­ma­tic ret­rie­val, ext­rac­tion, and sum­ma­riza­tion of know­ledge found in unstruc­tu­red text. In the bio­me­dical domain, sources of know­ledge-rich text include cli­nical docu­ments (e.g., cli­nical notes, pat­ho­logy reports) and publis­hed lite­ra­ture (e.g., jour­nal articles, cli­nical trial reports). Trans­for­ming this large volume of text into actio­nable know­ledge faci­li­ta­tes cli­nical acti­vi­ties such as cli­nical deci­sion sup­port, patient-co­hort identifica...