An introduction to mathematics of emerging biomedical imaging (Q930270)

This book consists of an introductive chapter, four parts, a bibliography, and an index. It does not contain a list of symbols, although there are many symbols in the book. The introductive chapter, Biomedical imaging modalities, contains a list of imaging procedures and a very short description of these procedures. Part 1, Mathematical tools, consists of two chapters, where a compilation of some known definitions and results from analysis and potential theory is given. Chapter 2. Preliminaries. Chapter 3. Layer potentials. Part 2, General reconstruction algorithms, consists of three chapters. Chapter 4. Tomographic imaging with non-diffracting sources. Chapter 5. Tomographic imaging with diffracting sources. Chapter 6. Biomagnetic source imaging. Part 3, Anomaly detection algorithms, consists of two chapters. Chapter 7. Small volume expansions. Chapter 8. Imaging techniques. Part 4, Hybrid imaging techniques, consists of three chapters. Chapter 9. Magnetic resonance electric impedance tomography. Chapter 10. Impediography. Chapter 11. Magnetic resonance elastography. Chapter 4 deals with tomography (CT=computerized tomography), and the presentation touches only upon some simple well-known results related to the Radon transform inversion. The ideas of local and pseudolocal tomography, developed by \textit{A. G. Ramm} and \textit{A. I. Katsevich} [The Radon transform and local tomography. Boca Raton, FL: CRC Press (1996; Zbl 0863.44001)] are not even mentioned. The author mentions very briefly the problem of summation of series with noisy coefficients, representing the reconstructed function, but does not give methods for stable summation of such series, and does not formulate known results on this problem [see, e.g., the book op. cit., p. 440]. Chapter 5 deals with impedance tomography (inverse conductivity problem) and with the Born inversion in ultrasound and microwave tomographies. The presentation here is very schematic, and basic difficulties are not mentioned. The reader is not told that although the Born approximation can be rigorously justified for weak perturbations and the error estimates are known, but the Born approximation for solving the inverse problem is always an ill-posed problem, no matter how weak the perturbation is [see, e.g., \textit{A.G. Ramm}, ``Is the Born approximation good for solving the inverse problem when the potential is small?'', J. Math. Anal. Appl. 147, No. 2, 480--485 (1990; Zbl 0705.35153)]. On p.110 the author assumes that the conductivity perturbation is small and vanishes on the boundary of the body. The last assumption is physically not justified. A recovery method which does not need this assumption is given in the monograph of \textit{A.G. Ramm}, ``Multidimensional inverse scattering problems.'' Longman/Wiley, N.Y. (1992; Zbl 0746.35056), pp. 72--80. The author gives no error estimates for the inversion formulas and does not discuss the case of noisy data. Chapter 6 deals with EEG (Electroencephalography) and MEG (Magnetoencephalography) problems. The presentation is again very schematic. Some assumptions are not explained (e.g., on p. 119 it is assumed that the solution to equation (6.3) has to satisfy the Neumann boundary condition and, in addition, its integral over the boundary should vanish). On p. 120 the Geselowitz formula is mentioned, but no reference is given. The formula below (6.6) is incorrect, in general, and it would help the reader if the assumption under which this formula is valid would be formulated explicitly on this page. In formula (6.8) the same symbol is used for the cross product and for the multiplication by a scalar. This may confuse the reader. On p. 121 the derivation of the basic formula is not given in detail. Chapter 7 deals with imaging of small inhomogeneities. The case of noisy data is not discussed. The notion of the polarizability tensor plays crucial role in this theory. The author did not mention that analytic formulas for polarizability tensors for homogeneous bodies of arbitrary shapes are derived in the monograph of \textit{A. G. Ramm}, ``Wave scattering by small bodies of arbitrary shapes.'' River Edge, NJ: World Scientific (2005; Zbl 1081.78001)]. He does not mention the well-known monograph of \textit{G. Polya} and \textit{G. Szegö}, ``Isoperimetric inequalities in mathematical physics.'' Princeton Univ. Press (1951; Zbl 0044.38301), where polarizability tensors are discussed. Chapter 8 deals with MUSIC algorithms and time-reversal imaging. The difficulties, arising when one uses these algorithms, are not discussed. The author recommends parameter fitting as a method for solving inverse problems. Although such a method is widely used by practitioners, one has to make clear to the practitioners that no error estimates are available for such a method, and that its errors, in general, can be arbitrary large. When one deals with medical imaging, it is highly desirable to have some degree of reliability of the inversion procedures. In Chapters 9-11 some hybrid imaging procedures are discussed very briefly. Sometimes one finds misprints in the book (e.g, p. 100, line 2; p. 154, l. 4), but they are not too bothersome. The book may be useful for non-specialists, but an in-depth presentation of the questions, touched upon in this book, is desirable. Students may use this book, but should be given warnings about difficulties one faces when inversion algorithms are applied practically.