000			a
999			_c31107 _d31107
008			221104b xxu||||| |||| 00| 0 eng d
020			_a9781316514993
082			_a610.285 _bPAP
100			_aPapademetris, Xenophon
245			_aIntroduction to medical software : foundations for digital health, devices, and diagnostics
260			_bCambridge University Press, _c2022 _aCambridge :
300			_axvi, 322 p. ; _bill., _c25 cm
365			_b57.99 _cGBP _d95.20
490			_aCambridge texts in biomedical engineering
504			_aIncludes bibliographical references and index.
520			_a"The goal of this book is to provide in one brief and accessible volume a survey of the critical material involved in the design, implementation, and management of medical software for both standalone software ("software as a medical device - SaMD") and software that is part of a physical medical device. One will find more detailed treatments of many of the topics covered in this book in specialized books that focus on some of the topics we cover (e.g. software engineering, systems engineering, probability theory, machine learning). Depth was not our goal; this book is explicitly designed to provide a broad survey. Our hope is to familiarize the reader with the span of topics he or she may need in entering this field and to provide pointers to more specialized publications as this becomes necessary. For example, most computer scientists have very limited exposure to statistical decision theory, and we think that even the cursory coverage in this book will at least enable them to understand "what they do not know" and seek help as opposed to being ignorant of this entire field and attempting to reinvent the wheel in an amateurish manner! An emerging challenge in medical software is the increasing use of big data and artificial intelligence/machine learning (AI/ML) techniques. This places an even greater stress on proper software design and management. Given that these are "black box" methods, in which the human understanding of what actually is going on is limited, a proper software quality process will be even more critical in creating reliable software tools. We introduce this topic in Section 1.3. In that section we also provide pointers to the other sections of the book in which we discuss issues related to the use of AI/ML methods. This is an introductory book. One can and should follow the material here with further study, using both original regulatory materials, industry standards,1 and more advanced books.2 Our goal can be summarized by the phrase "to convert unknown unknowns to known unknowns." Our goal is to make our reader aware of important material he or she is not as familiar with as one should be, and to pursue further study to acquire such knowledge. This is not a programming book. Our goal is to describe the enabling activities that support programmers in producing high-quality software in the context of medical applications. We are less concerned by questions such as 'How should we code?' Our focus, rather, is on answering higher-level questions such as 'How do we decide what we need to code?' and 'How should the process be organized?' There is a wealth of material available that describes the actual coding process, and, therefore, we chose not to duplicate this type of description here.
650			_aArtificial Intelligence
650			_aIntelligence artificielle
650			_aLogiciels
650			_aMedical Informatics
650			_aSoftware Validation
650			_a User need
650			_a Healthcare
650			_a Electronic health records
650			_aFDA
650			_aGDPR
650			_aMedical images
650			_aOff-the-shelf libraries
650			_aPatient
650			_aRisk management
650			_aSoftware deployment
650			_a Traceability
650			_aClinical trials
650			_a Quality management system
700			_aQuraishi, Ayesha N.
700			_aLicholai, Gregory P.
942			_2ddc _cBK