- Email: [email protected]
Clinical Usefulness of Three-Dimensional Reconstruction of the Temporal Bone from CT Scans in Cholesteatoma Cases
Auris·Nasus·Larynx (Tokyo) 20,167-173 (1993)
CLINICAL USEFULNESS OF THREE-DIMENSIONAL RECONSTRUCTION OF THE TEMPORAL BONE FROM CT SCANS IN CHOLESTEATOMA CASES Masahiro
KAWANA,
M.D. and Yuichi
NAKANO,
M.D.
Department of Otolaryngology, Niigata University School of Medicine. Niigata, Japan
The importance of computed tomography (CT) in analyzing temporal bone diseases has increased, and the ability to reconstruct the temporal bone structures in three dimensions from multiple CT films has been required. In order to facilitate the visualization of temporal bone structures, we tried reconstructing temporal bone CT images threedimensionally using a personal computer, and evaluated the possibility of using three-dimensional CT images clinically. Temporal bone CT scan films from five cases of cholesteatoma and a case of otosclerosis as control were examined. Four temporal bone structures (temporal bone contour, middle ear, mastoid, and inner ear) and cholesteatoma, extracted from enlarged black and white CT films, were inputted to the personal computer. Data import and image reconstruction procedures were performed using commercially available software. Our results indicate that threedimensional reconstructions contribute to visualizing temporal bone structures spatially, and to choosing surgical approaches in difficult cases, such as petrous bone cholesteatoma. In conclusion, three-dimensional reconstructions using a personal computer is useful in the diagnosis and treatment of cholesteatoma. Since its advent, computed tomography (CT) scanning has assumed a critical role in the development of otolaryngology. High-resolution CT is especially valuable in interpreting the anatomical structure of the temporal bone in detail,1 and to determine the position and characteristics of abnormalities, surgeons are forced to examine those many two-dimensional (2-D) CT films. If threedimensional (3-D) images reconstructed from such CT films were available, and if the reconstructed images could be observed from any angle or from inside, surgical interpretation of the abnormalities might be greatly facilitated and the new information might be helpful in deciding on the operative approach. Because the progress in 3-D graphics using personal computer has been surprisingly fast, Received for publication
December 3, 1992 167
M. KA WANA and Y. NAKANO
168
excellent software is available for personal use. We therefore sought to determine whether these 2-D CT films could be reconstructed three-dimensionally using a personal computer, and assessed their clinical usefulness. CASES AND METHODS
Five cases of cholesteatoma were used for this study. A case of otosclerosis was used as control. Table 1 shows the diagnosis of the cases and the number of the axial CT slices for 3-D reconstruction. The diagnosis and operations of all the cases were performed at the Department of Otolaryngology, Niigata University School of Medicine. Conventional 2-D CT scans were obtained for routine clinical indications, and individual scanning parameters were based upon the standard protocols of the Department of Radiology, Niigata University School of Medicine. Routine head immobilization was employed for all studies, and all CT scans were obtained on high-resolution CT scanners, the Siemens Somatom DR3 (Germany) with a 2 mm interval of scan circle, and the Hitachi WlOOO (Japan) with a 1.5 mm interval of scan circle. The acquired CT scans were all axial sections, based on the Reid's base line. No coronal section was performed in all cases. All studies were conducted retrospectively; consequently none of the patients were subjected to the risks of sedation or additional radiation exposure in order to facilitate 3-D reconstruction. CT films were enlarged to 24 X 19 cm of black and white prints from the original films. The contours of temporal bone shape, middle ear, inner ear, mastoid, and cholesteatoma were extracted from the films. The contours were traced using graphic digitizer (KL4300, Grafteck, Japan) and inputted into a personal computer operating system (PC9801RA, NEC, Japan). All traced data in the computer were connected section to section manually on the computer display, and ultimately, the data were reconstructed three-dimensionally (Fig. 1). All procedures were performed with 3-D graphic software (Cosmozone 2SA, Nikon, Japan). When reconstructing the temporal bone, several contours, a maximum of five, can be chosen and drawn in the display. In addition, the reconstructed image can be processed in several ways: (i) the image can be rotated and cut, (ii) makeup and Table I.
Diagnosis of the cases and number of CT slices for 3-D reconstruction.
Case
Age
Sex
Diagnosis
Number of CT slices
I
34 16 40 46 36 16
F M M F M M
Otosclerosis (control) Petrous bone cholesteatoma Petrous bone cholesteatoma Petro us bone cholesteatoma Petrous bone cholesteatoma Attic cholesteatoma (extended to mastoid)
8 9
2 3 4 5 6
M: male, F: female.
27 13 14 18
3-D RECONSTRUCTION OF CHOLESTEATOMA
169
color of the image can be changed. In the final reconstructed image, the temporal bone outline is colored blue, the middle ear yellow, the inner ear red, the mastoid green, and the cholesteatoma white. RESULTS
Figure 2 is the 3-D image of Case 1, viewed supero-posteriorly. The reconstructed components facilitate understanding the relationships of each structures much more than regular 2-D CT films. In the CT films of Case 2, a uniform soft tissue density can be seen in the petro us bone and inner ear (Fig. 3). Of special interest is the destruction of the cochlea on the petrosal side. Figure 4 is a reconstructed image from nine CT films from Case 2. The image is a supero-posterior view. A cholesteatoma can be seen in the petro us bone as well as the destroyed medial half of the inner ear. Viewed from postero-Iaterally, which is a direction of surgical approach, it is apparent that removal of the cholesteatoma with preservation of the remaining inner ear would be difficult (Fig. 5). In Case 3, the CT films showed that the mastoid was fully occupied by cholesteatoma, and the posterior wall of the mastoid has been destroyed (Fig. 6). Inner ear destruction, however, is unclear. The reconstructed image from 27 CT films tells that the cholesteatoma originated in the petrous bone and expanded into the mastoid cavity, so that the remaining inner ear structure is completely enveloped by cholesteatoma (Fig. 7). Case 4 is an extremely difficult case to plan the surgical approach to the cholesteatoma from the middle ear on the basis of CT films. The 3-D image clearly demonstrated inter-relationships between the mastoid, inner ear, and cholesteatoma. In Case 5, severe bone destruction can be seen in anterior wall of the temporal bone in CT films, but to understand the size and position of the cholesteatoma was difficult. By observing the 3-D image, it was easy to realize that the main portion of the cholesteatoma was in the middle cranial fossa. In Case 6, CT films demonstrated severe destruction of the posterior wall of the mastoid, but the relationship of the cholesteatoma and inner ear was not clear. The 3-D image clearly showed that the large cholesteatoma invaded to the middle and inner ear, as confirmed by this surgical findings in which the dura and sigmoid sinus were exposed, and fistulas of the all three semicircular canals were present. DISCUSSION
Recently, 3-D reconstructions from 2-D image data, such as CT films, have become popular in clinical diagnosis. Three-dimensional reconstructions of the facial-mandibular bone area, in particular, often contribute to diagnosis and surgical planning. 2.3
170
M. KA W ANA and Y. NAKANO
In the temporal bone area, most physicians recognize that high-resolution CT and MRI have greatly contributed to the diagnosis and treatment, I and obtaining such image data has become commonplace in routine clinical work. Consequently, otologists have long struggled to grasp complicated structure of the temporal bone spatially, but that is not easy, even for specialists in otological radiology. Threedimensional reconstruction of temporal bone structures from CT or MRI films using computers has been attempted to facilitate interpretation of 2-D image data. This trend is based on the results of 2-D reconstruction from serial histological sections of organs. For example in otology, several investigators have reported 3-D reconstruction of the eustachian tube, temporal bone, cochlea, vestibular organs, and endolymphatic sac, which assist in better understanding their histological structures. 4•9 A computer system, associated instruments for image data input, and software are most important materials required for 3-D reconstruction. To reconstruct 3-D images from histological sections, the use of small computers and hand tracing of the sections are common in reconstructing 3-D images from histological sections.4•7•9 As rapid improvements have been made in computers, reconstruction techniques using computers in CT systems with direct incorporation of the scanning data and automatic creation of 3-D images have been assessed for clinical use, and reconstruction of the temporal bone from CT scans using these systems has been attempted by several investigators. IO• l2 Other investigators have tried to acquire more clear pictures using specially authorized software and hardware. 13 Comparison between direct CT data incorporation and hand tracing of the temporal bone CT films using personal computer concluded that direct CT data incorporation created smoother 3-D image contours than hand tracing. 14 We use personal computers, hand tracing of the CT films, and commercially available software, because (i) automatic reconstruction systems associated with CTs are still very expensive, making such system unpopular, (ii) hand tracing of the CT films is necessary to select individual contours of temporal bone structures freely .
Fig. I. One of the processes of the 3·D reconstruction on the computer display. Each component of the CT slices was connected manually. Fig. 2. Reconstructed image from Case I (control), viewed from postero-superiorly. Blue, bone outline; yellow, middle ear; green, mastoid; red, inner ear. Fig. 3. A slice of axial CT of Case 2. A soft tissue density can be seen in the petrous bone. Fig. 4. Reconstructed image from Case 2, viewed from postero-superiorly. Cholesteatoma (white) is present in the petrous bone. Fig. 5. Reconstructed image from Case 2, viewed from postero-Iaterally. After rotation, the outer part of the image is removed. Fig. 6. CT film from Case 3. The mastoid is fully occupied by cholesteatoma, and the posterior wall of the mastoid has been destroyed. Fig. 7. Reconstructed image of Case 3. The outer part of the temporal bone (blue cross-hatching) has been removed. The inner ear (red) is enveloped in the cholesteatoma (white).
3-D RECONSTRUCTION OF CHOLESTEATOMA
17 1
172
M. KAWANA and Y. NAKANO
As shown in the cases described, the features of our 3-D reconstruction are: (i) several, a maximum of five, temporal bone components can be reconstructed at once; (ii) by using a rotation and cutting technique on the reconstructed images, it is possible to observe the objects from angles impossible to obtain with CT scans; and (iii) the process of reconstruction itself helps to increase our CT reading skills. Among our results, 3-D reconstruction of the temporal bone from CT films of Case 2 to 5, difficult cases of the temporal bone cholesteatoma, are most valuable for understanding the position of the cholesteatoma and the areas of bone destruction. Petrous bone cholesteatoma is one of the most difficult lesions to diagnose and to plan surgery for. 3-D reconstructions using computer systems allow us to our misty reconstructions in our mind, and provide us with direct and exact images of changes in the temporal bone. The points to improve in our system are: (i) to reconstruct detailed structures in the temporal bone, such as the ossicular chains, (ii) to reconstruct threedimensionally at real-time during CT operations. Increases in the resolving power of the CT system and determining optimum conditions during CT examination as an aid to 3-D reconstruction may possibly resolve the former point, and future development of computer systems may also resolve the latter point. In conclusion, 3-D reconstruction using computer system is very helpful in understanding the 3-D anatomy of temporal bone, and contributes to the treatment of difficult cholesteatoma cases. REFERENCES I. 2. 3. 4. 5. 6. 7. 8. 9.
10.
Olson IE, Dorwart RH, Brant WE: Use of high resolution thin section CT scanning of the petrous bone in temporal bone anomalies. Laryngoscope 92: 1274-1278, 1982. Moaddab MB, Dumas AL, Chavoor AG, et al: Temporomandibular joint: Computed tomographic three-dimensional reconstruction. Am I Orthod 88:342-352, 1985. Marsh IL, Baca D, Vannier MW: Facial musculoskeletal asymmetry in hemifacial microsomia. Cleft Palate I 26:292-302, 1989. Mori K, Naito Y, Hirono Y, et al: Three-dimensional computer graphics of the eustachian tube. Am I Otolaryngol 8:211-213, 1987. Harada T, Ishii S, Tayama N: Three-dimensional reconstruction of the temporal bone from histologic sections. Arch Otolaryngol Head Neck Surg 114:1139-1142, 1988. Green ID, Marion MS, Erickson BJ, et al: Three-dimensional reconstruction of the temporal bone. Laryngoscope 100: 1-4, 1990. Takagi A, Sando I: Computer-aided three-dimensional reconstruction and measurement of the vestibular end-organs. Otolaryngol Head Neck Surg 88:195-202, 1988. Linthicum FH, Galey FR: Computer-aided reconstruction of the endolymphatic sac. Acta Otolaryngol 91 :423-429, 1981. Antunez JCM, Galey FR, Linthicum FH, et al: Computer-aid and graphic reconstruction of the human endolymphatic duct and sac. A method for comparing Meniere's and non-Meniere's disease cases. Ann Otol Rhinol Laryngol 89(Suppl 76):23-32, 1980. LaRouere MJ, Niparko JK, Gebarski SS, et al: Three-dimensional x-ray computed tomography of the temporal bone as an aid to surgical planning. Otolaryngol Head Neck Surg 103:740-747, 1990.
3-D RECONSTRUCTION OF CHOLESTEATOMA II.
12. 13.
14.
173
Yamamoto E, Mizukami C, !sono M, et al: Observation of the external aperture of the vestibular aqueduct using three-dimensional surface reconstruction imaging. Laryngoscope 101:480-483, 1991. Brogan M, Chakeres DW, Schmalbrock P: High-resolution 3DFT-MR imaging of the endolymphatic duct and soft tissues of the otic capsule. ANJR 12:1-11, 1991. Davis RE, Levoy M, Rosenman JG, et al: Three-dimensional high-resolution volume rendering (HRVR) of computed tomography data: applications to otolaryngology-head and neck surgery. Laryngoscope 101:573-582, 1991. Seldon HL: Three-dimensional reconstruction of temporal bone from computed tomographic scans on a personal computer. Arch Otolaryngol Head Neck Surg 117:1158-1161, 1991.
Request reprints to:
Dr. M. Kawana, Department of Otolaryngology, Niigata University School of Medicine, Asahimachi-I, Niigata 951, Japan
CLINICAL USEFULNESS OF THREE-DIMENSIONAL RECONSTRUCTION OF THE TEMPORAL BONE FROM CT SCANS IN CHOLESTEATOMA CASES Masahiro
KAWANA,
M.D. and Yuichi
NAKANO,
M.D.
Department of Otolaryngology, Niigata University School of Medicine. Niigata, Japan
The importance of computed tomography (CT) in analyzing temporal bone diseases has increased, and the ability to reconstruct the temporal bone structures in three dimensions from multiple CT films has been required. In order to facilitate the visualization of temporal bone structures, we tried reconstructing temporal bone CT images threedimensionally using a personal computer, and evaluated the possibility of using three-dimensional CT images clinically. Temporal bone CT scan films from five cases of cholesteatoma and a case of otosclerosis as control were examined. Four temporal bone structures (temporal bone contour, middle ear, mastoid, and inner ear) and cholesteatoma, extracted from enlarged black and white CT films, were inputted to the personal computer. Data import and image reconstruction procedures were performed using commercially available software. Our results indicate that threedimensional reconstructions contribute to visualizing temporal bone structures spatially, and to choosing surgical approaches in difficult cases, such as petrous bone cholesteatoma. In conclusion, three-dimensional reconstructions using a personal computer is useful in the diagnosis and treatment of cholesteatoma. Since its advent, computed tomography (CT) scanning has assumed a critical role in the development of otolaryngology. High-resolution CT is especially valuable in interpreting the anatomical structure of the temporal bone in detail,1 and to determine the position and characteristics of abnormalities, surgeons are forced to examine those many two-dimensional (2-D) CT films. If threedimensional (3-D) images reconstructed from such CT films were available, and if the reconstructed images could be observed from any angle or from inside, surgical interpretation of the abnormalities might be greatly facilitated and the new information might be helpful in deciding on the operative approach. Because the progress in 3-D graphics using personal computer has been surprisingly fast, Received for publication
December 3, 1992 167
M. KA WANA and Y. NAKANO
168
excellent software is available for personal use. We therefore sought to determine whether these 2-D CT films could be reconstructed three-dimensionally using a personal computer, and assessed their clinical usefulness. CASES AND METHODS
Five cases of cholesteatoma were used for this study. A case of otosclerosis was used as control. Table 1 shows the diagnosis of the cases and the number of the axial CT slices for 3-D reconstruction. The diagnosis and operations of all the cases were performed at the Department of Otolaryngology, Niigata University School of Medicine. Conventional 2-D CT scans were obtained for routine clinical indications, and individual scanning parameters were based upon the standard protocols of the Department of Radiology, Niigata University School of Medicine. Routine head immobilization was employed for all studies, and all CT scans were obtained on high-resolution CT scanners, the Siemens Somatom DR3 (Germany) with a 2 mm interval of scan circle, and the Hitachi WlOOO (Japan) with a 1.5 mm interval of scan circle. The acquired CT scans were all axial sections, based on the Reid's base line. No coronal section was performed in all cases. All studies were conducted retrospectively; consequently none of the patients were subjected to the risks of sedation or additional radiation exposure in order to facilitate 3-D reconstruction. CT films were enlarged to 24 X 19 cm of black and white prints from the original films. The contours of temporal bone shape, middle ear, inner ear, mastoid, and cholesteatoma were extracted from the films. The contours were traced using graphic digitizer (KL4300, Grafteck, Japan) and inputted into a personal computer operating system (PC9801RA, NEC, Japan). All traced data in the computer were connected section to section manually on the computer display, and ultimately, the data were reconstructed three-dimensionally (Fig. 1). All procedures were performed with 3-D graphic software (Cosmozone 2SA, Nikon, Japan). When reconstructing the temporal bone, several contours, a maximum of five, can be chosen and drawn in the display. In addition, the reconstructed image can be processed in several ways: (i) the image can be rotated and cut, (ii) makeup and Table I.
Diagnosis of the cases and number of CT slices for 3-D reconstruction.
Case
Age
Sex
Diagnosis
Number of CT slices
I
34 16 40 46 36 16
F M M F M M
Otosclerosis (control) Petrous bone cholesteatoma Petrous bone cholesteatoma Petro us bone cholesteatoma Petrous bone cholesteatoma Attic cholesteatoma (extended to mastoid)
8 9
2 3 4 5 6
M: male, F: female.
27 13 14 18
3-D RECONSTRUCTION OF CHOLESTEATOMA
169
color of the image can be changed. In the final reconstructed image, the temporal bone outline is colored blue, the middle ear yellow, the inner ear red, the mastoid green, and the cholesteatoma white. RESULTS
Figure 2 is the 3-D image of Case 1, viewed supero-posteriorly. The reconstructed components facilitate understanding the relationships of each structures much more than regular 2-D CT films. In the CT films of Case 2, a uniform soft tissue density can be seen in the petro us bone and inner ear (Fig. 3). Of special interest is the destruction of the cochlea on the petrosal side. Figure 4 is a reconstructed image from nine CT films from Case 2. The image is a supero-posterior view. A cholesteatoma can be seen in the petro us bone as well as the destroyed medial half of the inner ear. Viewed from postero-Iaterally, which is a direction of surgical approach, it is apparent that removal of the cholesteatoma with preservation of the remaining inner ear would be difficult (Fig. 5). In Case 3, the CT films showed that the mastoid was fully occupied by cholesteatoma, and the posterior wall of the mastoid has been destroyed (Fig. 6). Inner ear destruction, however, is unclear. The reconstructed image from 27 CT films tells that the cholesteatoma originated in the petrous bone and expanded into the mastoid cavity, so that the remaining inner ear structure is completely enveloped by cholesteatoma (Fig. 7). Case 4 is an extremely difficult case to plan the surgical approach to the cholesteatoma from the middle ear on the basis of CT films. The 3-D image clearly demonstrated inter-relationships between the mastoid, inner ear, and cholesteatoma. In Case 5, severe bone destruction can be seen in anterior wall of the temporal bone in CT films, but to understand the size and position of the cholesteatoma was difficult. By observing the 3-D image, it was easy to realize that the main portion of the cholesteatoma was in the middle cranial fossa. In Case 6, CT films demonstrated severe destruction of the posterior wall of the mastoid, but the relationship of the cholesteatoma and inner ear was not clear. The 3-D image clearly showed that the large cholesteatoma invaded to the middle and inner ear, as confirmed by this surgical findings in which the dura and sigmoid sinus were exposed, and fistulas of the all three semicircular canals were present. DISCUSSION
Recently, 3-D reconstructions from 2-D image data, such as CT films, have become popular in clinical diagnosis. Three-dimensional reconstructions of the facial-mandibular bone area, in particular, often contribute to diagnosis and surgical planning. 2.3
170
M. KA W ANA and Y. NAKANO
In the temporal bone area, most physicians recognize that high-resolution CT and MRI have greatly contributed to the diagnosis and treatment, I and obtaining such image data has become commonplace in routine clinical work. Consequently, otologists have long struggled to grasp complicated structure of the temporal bone spatially, but that is not easy, even for specialists in otological radiology. Threedimensional reconstruction of temporal bone structures from CT or MRI films using computers has been attempted to facilitate interpretation of 2-D image data. This trend is based on the results of 2-D reconstruction from serial histological sections of organs. For example in otology, several investigators have reported 3-D reconstruction of the eustachian tube, temporal bone, cochlea, vestibular organs, and endolymphatic sac, which assist in better understanding their histological structures. 4•9 A computer system, associated instruments for image data input, and software are most important materials required for 3-D reconstruction. To reconstruct 3-D images from histological sections, the use of small computers and hand tracing of the sections are common in reconstructing 3-D images from histological sections.4•7•9 As rapid improvements have been made in computers, reconstruction techniques using computers in CT systems with direct incorporation of the scanning data and automatic creation of 3-D images have been assessed for clinical use, and reconstruction of the temporal bone from CT scans using these systems has been attempted by several investigators. IO• l2 Other investigators have tried to acquire more clear pictures using specially authorized software and hardware. 13 Comparison between direct CT data incorporation and hand tracing of the temporal bone CT films using personal computer concluded that direct CT data incorporation created smoother 3-D image contours than hand tracing. 14 We use personal computers, hand tracing of the CT films, and commercially available software, because (i) automatic reconstruction systems associated with CTs are still very expensive, making such system unpopular, (ii) hand tracing of the CT films is necessary to select individual contours of temporal bone structures freely .
Fig. I. One of the processes of the 3·D reconstruction on the computer display. Each component of the CT slices was connected manually. Fig. 2. Reconstructed image from Case I (control), viewed from postero-superiorly. Blue, bone outline; yellow, middle ear; green, mastoid; red, inner ear. Fig. 3. A slice of axial CT of Case 2. A soft tissue density can be seen in the petrous bone. Fig. 4. Reconstructed image from Case 2, viewed from postero-superiorly. Cholesteatoma (white) is present in the petrous bone. Fig. 5. Reconstructed image from Case 2, viewed from postero-Iaterally. After rotation, the outer part of the image is removed. Fig. 6. CT film from Case 3. The mastoid is fully occupied by cholesteatoma, and the posterior wall of the mastoid has been destroyed. Fig. 7. Reconstructed image of Case 3. The outer part of the temporal bone (blue cross-hatching) has been removed. The inner ear (red) is enveloped in the cholesteatoma (white).
3-D RECONSTRUCTION OF CHOLESTEATOMA
17 1
172
M. KAWANA and Y. NAKANO
As shown in the cases described, the features of our 3-D reconstruction are: (i) several, a maximum of five, temporal bone components can be reconstructed at once; (ii) by using a rotation and cutting technique on the reconstructed images, it is possible to observe the objects from angles impossible to obtain with CT scans; and (iii) the process of reconstruction itself helps to increase our CT reading skills. Among our results, 3-D reconstruction of the temporal bone from CT films of Case 2 to 5, difficult cases of the temporal bone cholesteatoma, are most valuable for understanding the position of the cholesteatoma and the areas of bone destruction. Petrous bone cholesteatoma is one of the most difficult lesions to diagnose and to plan surgery for. 3-D reconstructions using computer systems allow us to our misty reconstructions in our mind, and provide us with direct and exact images of changes in the temporal bone. The points to improve in our system are: (i) to reconstruct detailed structures in the temporal bone, such as the ossicular chains, (ii) to reconstruct threedimensionally at real-time during CT operations. Increases in the resolving power of the CT system and determining optimum conditions during CT examination as an aid to 3-D reconstruction may possibly resolve the former point, and future development of computer systems may also resolve the latter point. In conclusion, 3-D reconstruction using computer system is very helpful in understanding the 3-D anatomy of temporal bone, and contributes to the treatment of difficult cholesteatoma cases. REFERENCES I. 2. 3. 4. 5. 6. 7. 8. 9.
10.
Olson IE, Dorwart RH, Brant WE: Use of high resolution thin section CT scanning of the petrous bone in temporal bone anomalies. Laryngoscope 92: 1274-1278, 1982. Moaddab MB, Dumas AL, Chavoor AG, et al: Temporomandibular joint: Computed tomographic three-dimensional reconstruction. Am I Orthod 88:342-352, 1985. Marsh IL, Baca D, Vannier MW: Facial musculoskeletal asymmetry in hemifacial microsomia. Cleft Palate I 26:292-302, 1989. Mori K, Naito Y, Hirono Y, et al: Three-dimensional computer graphics of the eustachian tube. Am I Otolaryngol 8:211-213, 1987. Harada T, Ishii S, Tayama N: Three-dimensional reconstruction of the temporal bone from histologic sections. Arch Otolaryngol Head Neck Surg 114:1139-1142, 1988. Green ID, Marion MS, Erickson BJ, et al: Three-dimensional reconstruction of the temporal bone. Laryngoscope 100: 1-4, 1990. Takagi A, Sando I: Computer-aided three-dimensional reconstruction and measurement of the vestibular end-organs. Otolaryngol Head Neck Surg 88:195-202, 1988. Linthicum FH, Galey FR: Computer-aided reconstruction of the endolymphatic sac. Acta Otolaryngol 91 :423-429, 1981. Antunez JCM, Galey FR, Linthicum FH, et al: Computer-aid and graphic reconstruction of the human endolymphatic duct and sac. A method for comparing Meniere's and non-Meniere's disease cases. Ann Otol Rhinol Laryngol 89(Suppl 76):23-32, 1980. LaRouere MJ, Niparko JK, Gebarski SS, et al: Three-dimensional x-ray computed tomography of the temporal bone as an aid to surgical planning. Otolaryngol Head Neck Surg 103:740-747, 1990.
3-D RECONSTRUCTION OF CHOLESTEATOMA II.
12. 13.
14.
173
Yamamoto E, Mizukami C, !sono M, et al: Observation of the external aperture of the vestibular aqueduct using three-dimensional surface reconstruction imaging. Laryngoscope 101:480-483, 1991. Brogan M, Chakeres DW, Schmalbrock P: High-resolution 3DFT-MR imaging of the endolymphatic duct and soft tissues of the otic capsule. ANJR 12:1-11, 1991. Davis RE, Levoy M, Rosenman JG, et al: Three-dimensional high-resolution volume rendering (HRVR) of computed tomography data: applications to otolaryngology-head and neck surgery. Laryngoscope 101:573-582, 1991. Seldon HL: Three-dimensional reconstruction of temporal bone from computed tomographic scans on a personal computer. Arch Otolaryngol Head Neck Surg 117:1158-1161, 1991.
Request reprints to:
Dr. M. Kawana, Department of Otolaryngology, Niigata University School of Medicine, Asahimachi-I, Niigata 951, Japan