An algorithm for tracking microcatheters in fluoroscopy.

Currently, a large number of endovascular interventions are performed for treatment of intracranial aneurysms. For these treatments, correct positioning of microcatheter tips, microguide wire tips, or coils is essential. Techniques to detect such devices may facilitate endovascular interventions. In this paper, we describe an algorithm for tracking of microcatheter tips during fluoroscopically guided neuroendovascular interventions. A sequence of fluoroscopic images (1,024 x 1,024 x 12 bits) was acquired using a C-arm angiography system as a microcatheter was passed through a carotid phantom which was on top of a head phantom. The carotid phantom was a silicone cylinder containing a simulated vessel with the shape and curvatures of the internal carotid artery. The head phantom consisted of a human skull and tissue-equivalent material. To detect the microcatheter in a given fluoroscopic frame, a background image consisting of an average of the four previous frames is subtracted from the current frame, the resulting image is filtered using a matched filter, and the position of maximum intensity in the filtered image is taken as the catheter tip position in the current frame. The distance between the tracked position and the correct position (error distance) was measured in each of the fluoroscopic images. The mean and standard deviation of the error distance values were 0.277 mm (1.59 pixels) and 0.26 mm (1.5 pixels), respectively. The error distance was less than 3 pixels in the 93.0% frames. Although the algorithm intermittently failed to correctly detect the catheter, the algorithm recovered the catheter in subsequent frames.

Tracking technique of a micro guide wire in sequential fluorograms.

In this paper, we propose a tracking technique for a micro guide wire in a sequence of fluorograms. A technique in which the region-growing technique was embedded was developed. It gives the center line of a micro guide wire. A sequence of 1,024 x 1,024 x 16 bit fluorograms (111 frames) with a carotid phantom and head phantom was obtained with a CAS-8000V (Toshiba America Medical Systems, Inc., CA, USA) C-arm angiography system and a Hi-Torque Standard micro guide wire (Advanced Cardiovascular Systems, Inc., CA, USA). To evaluate the technique, we manually traced the guide wire in each sequence frame three times, and a "true" single-width micro guide wire was created from them. The number of pixels on the true guide wire and inside the two-pixel tolerance of the center line was counted in each fluorogram, and the percentage of that count based on the number of all pixels on the true guide wire was calculated as true positive (TP). In addition, the number of pixels on the center line and outside the two-pixel tolerance of the true guide wire was counted in each frame as false positive (FP). The tracking technique has a mean TP of 94.8% and a mean FP of 5.1 pixels/frame. Several frames have low TP and high FP, but the technique could continue to track the micro guide wire until the end of the sequence. We therefore concluded that we had developed an accurate automatic tracking technique for micro guide wires in fluoroscopic sequences.

Workflow analysis of medical staff in surgical wards based on time-motion study data.

The aim of this study is to clarify the change over time in the elements of work (job elements) and their features, as well as the relationship between job elements and the type of job, job class, and the role of the hospital they are performed in. A time-motion study was conducted on the medical staff in the surgical wards of two hospitals. An analysis of roles bys by (a) type or class of job type, and (b) hospital function was conducted. The number of working hours was analyzed, as well as the ratio of working hours with respect to direct and indirect job elements. The job elements required for each medical staff member were proven to differ by type of job (doctors and nurses) and also by job class (nurse leaders and staff). When comparing between hospital the differences in job elements were proven not to be a result of differences in hospital function, but to result from the ward system (ward design and nursing system).

Analysis of respiration-related movement of upper abdominal arteries: preliminary measurement for the development of a respiratory motion compensation technique of roadmap navigation.

In this study, we analyzed the respiratory motion of the upper abdominal arteries preliminary to developing a method of respiratory motion correction for the roadmap technique used in vascular interventions. We retrospectively obtained six digital angiography sequences taken during respiration. The levels of the right and left hemidiaphragms and the positions of artery bifurcation points were measured manually through each sequence. Artery bifurcation points were classified as the hepatic artery group (HAG), splenic artery group (SAG), and celiac group (CG). Correlations between the motions of each hemidiaphragm and of the artery bifurcation points in each group were determined. We found that the vertical motion of the HAG and CG matched that of the right hemidiaphragm (r = 0.924 and r = 0.888, respectively). The vertical motion of the SAG matched that of the left hemidiaphragm (r = 0.949). The mean horizontal movements for all groups were up to 1.90 mm. The vertical motion for each group matched that of the right or the left hemidiaphragm. These findings will facilitate the development of a method of respiratory motion correction for the roadmap technique.

[Usefulness of the classification technique of cerebral artery for 2D/3D registration.].

Several papers have proposed 2D/3D registration methods of the cerebral artery using magnetic resonance angiography (MRA) and digital subtraction angiography (DSA). Since differences between vessels in a DSA image and MRA volume data cause registration failure, we previously proposed a method to extract vessels from MRA volume data using a technique based on classification of the cerebral artery. In this paper, we evaluated the usefulness of this classification technique by evaluating the reliability of this 2D/3D registration method. This classification method divides the cerebral artery in MRA volume data into 12 segments. According to the results of the classification, structures corresponding to vessels on a DSA image can then be extracted. We applied the 2D/ 3D registration with /without classification to 16 pairs of MRA volume data and DSA images obtained from six patients. The registration results were scored into four levels (Excellent, Good, Fair and Poor). The rates of successful registration (> fair) were 37.5% for registration without classification and 81.3% for that with classification. These findings suggested that there was a low percentage of incorrectly extracted voxels and we could facilitate reliable registration. Thus, the classification technique was shown to be useful for feature-based 2D/ 3D registration.

Microcatheter tip enhancement in fluoroscopy: a comparison of techniques.

We compared three techniques for enhancement of microcatheter tips in fluoroscopic images: conventional subtraction technique (CST); averaged image subtraction technique (AIST), which we have developed; and double average filtering (DAF) technique, which uses nonlinear background estimates. A pulsed fluoroscopic image sequence was obtained as a microcatheter was passed through a carotid phantom that was on top of a head phantom. The carotid phantom was a silicone cylinder containing a simulated vessel with the shape and curvatures of the internal carotid artery. The three techniques were applied to the images of the sequence, then the catheter tip was manually identified in each image, and 100 x 100 pixel images, centered at the indicated microcatheter tip positions, were extracted for the evaluations. The signal-to-noise ratio (SNR) was calculated in each of the extracted images from which the mean value of the SNR and its standard deviation (SD) were calculated for each technique. The mean values and the standard deviations were 4.36 (SD 3.40) for CST, 6.34 (SD 3.62) for AIST, and 3.55 (SD 1.27) for DAF. AIST had a higher SNR compared to CST in almost all frames. Although DAF yielded the smallest mean SNR value, it yielded the best SNR in those frames in which the microcatheter tip did not move between frames. We conclude that AIST provides the best SNR for a moving microcatheter tip and that DAF is optimal for a temporarily stationary microcatheter tip.

Automatic anatomical labeling method of cerebral arteries in MR-angiography data set.

To improve the accuracy and robustness of 2D/3D registration of digital subtraction angiography images and magnetic resonance angiography (MRA) data, we have developed an automatic method for anatomical labeling of the cerebral arteries in MRA data. The anatomical labeling method is a location-based method which segments an artery tree to branches and classifies the branches into labeled segments, i.e., internal carotid arteries (ICA), basilar artery (BA), middle cerebral arteries (MCA), A1 segments of the anterior cerebral artery (ACA(A1)), other segments of the anterior cerebral artery (ACA), posterior communication arteries (PcomA) and posterior cerebral arteries (PCA), according to their location. Arteries were extracted from MRA data for this labeling method by the region-growing technique. Fifteen cases were examined to evaluate the method accuracy. The number of correctly segmented voxels in each artery segment was determined, and the correct labeling percentage was calculated based on the total number of voxels of the artery. Mean percentages were as follows: ACA, 82.7%; Right (R-) ACA(A1), 47.1%; Left (L-) ACA(A1), 46.1%; R-MCA, 80.4%; L-MCA, 74.1%; R-PcomA, 0.0%; L-PcomA, 3.3%; R-PCA, 60.3%; LPCA, 66.9%; R-ICA, 90.7%; L-ICA, 90.7%; BA, 89.9%; and total arteries, 84.1%. The ACA, MCA, ICA and BA were consistently identified correctly.

Methodology development for quantitative optimization of security enhancement in medical information systems -Case study in a PACS and a multi-institutional radiotherapy database-.

The target of our study is to establish the methodology for analyzing level of security requirements, for searching suitable security measures and for optimizing security distribution to every portion of medical practice. Quantitative expression must be introduced to our study as possible for the purpose of easy follow up of security procedures and easy evaluation of security outcomes or results. Results of system analysis by fault tree analysis (FTA) clarified that subdivided system elements in detail contribute to much more accurate analysis. Such subdivided composition factors very much depended on behavior of staff, interactive terminal devices, kinds of service, and routes of network. As conclusion, we found the methods to analyze levels of security requirements for each medical information systems employing FTA, basic events for each composition factor and combination of basic events. Methods for searching suitable security measures were found. Namely risk factors for each basic event, number of elements for each composition factor and candidates of security measure elements were found. Method to optimize the security measures for each medical information system was proposed. Namely optimum distribution of risk factors in terms of basic events were figured out, and comparison of them between each medical information systems became possible.

Evaluation of the effect of varying MPEG-2 compression ratios on digital coronary angiographic assessment of stenosis severity.

The purpose of this study is to evaluate the influence of MPEG-2 compression scheme on coronary angiography and to search the highest compression ratio at which no significant effect to accuracy of assessment of stenosis severity occurs. Forty-Four digital cine angiographies were used. Three cardiologists participated in a subjective study in which they read both uncompressed images and compressed images. Furthermore, an objective study was carried out to measure vessel stenosis ratio by using software. The influence of compression was evaluated by kappa statistics in case of subjective study and by both systematic error and random error in case of objective study. Kappa statistics between uncompressed image and compressed image at a ratio of 80:1 was significantly lower than that of other compression ratios such as 40:1. Similar results were obtained in objective evaluation. In this report, the authors provide the baseline for further studies on observer performance for motion images.

An inductive method for automatic generation of referring physician prefetch rules for PACS.

To prefetch images in a hospital-wide picture archiving and communication system (PACS), a rule must be devised to permit accurate selection of examinations in which a patient's images are stored. We developed an inductive method to compose prefetch rules from practical data which were obtained in a hospital using a decision tree algorithm. Our methods were evaluated on data acquired in Osaka University Hospital for one month. The data collected consisted of 58,617 cases of consultation reservations, 643,797 examination histories of patients, and 323,993 records of image requests in PACS. Four parameters indicating whether the images of the patient were requested or not for each consultation reservation were derived from the database. As a result, the successful selection sensitivity for consultations in which images were requested was approximately 0.8, and the specificity for excluding consultations accurately where images were not requested was approximately 0.7.

Development of a system which automatically acquires optimal discrete-valued attributes by dividing and grouping continuous-valued attributes to assist clinical decision making in radiotherapy.

The purposes of this study were first to develop a system which statistically tests results of radiotherapy and which automatically acquires an optimal discrete-valued attribute by dividing and grouping continuous-valued attributes, and second to find the optimal range of values such as tumor dose by taking account of the conditions and statistics in ROGAD (Radiation Oncology Greater Area Database), a multi-institutional database in Japan. Our ultimate goal is to assist clinical decision making for every patient. In this research, two algorithms for acquiring a boundary value were developed without detecting false boundaries or accidental errors of acquired boundaries. The resolution of detected discrete-valued attributes and speed of convergence were confirmed to be practical. The optimal range of given tumor dose with the best reaction and with the fewest complications is expected to be clarified.

Development of a Virtual Private Database for a multi-institutional Internet-based Radiation Oncology Database overcoming differences in protocols.

A multi-institutional Radiation Oncology Greater Area Database (ROGAD) was started in 1991 under the direction of the Japanese Society for Therapeutic Radiology and Oncology (JASTRO). Use of ROGAD was intended to allow reflection of results of data analysis into treatment strategy and treatment planning for individual cases, to provide quality assurance, to maximize the efficacy of radiotherapy, to allow assessment of new technologies or new modalities, and to optimize medical decision making. ROGAD collected 13,448 radiotherapy treatment cases from 325 facilities during the period from 1992 to 2001. In 2000, questionnaires were sent to 725 radiotherapy facilities throughout Japan, to further obtain the situation of the radiation oncology database. Workers at 179 facilities replied that"the protocol of my facility is different from ROGAD protocol and I must send data according to the ROGAD protocol". So, we developed the Virtual Private Database System (VPDS) which is operated as if an oncologist had a database solely owned by his own facility, in spite of actually operating ROGAD. VPDS realizes integration of different plural databases, regardless of differences in entry methods, protocols, definitions and interpretations of contents of clinical data elements between facilities.