Search

BACKGROUND: Radiologists widely use the minimum detectable difference (MDD) concept for inspecting the imaging quality and quantify the spatial resolution of scans. OBJECTIVE: This study adopted Taguchi’s dynamic algorithm to optimize the MDD of cardiac CT angiography (CTA) using a V-shaped line gauge and three PMMA phantoms (50, 70, and 90 kg). METHODS: The phantoms were customized in compliance with the ICRU-48 report, whereas the V-shaped line gauge was indigenous to solidify the cardiac CTA scan image quality by two adjacent peaks along the V-shaped slit. Accordingly, the six factors A-F assigned in this study were A (kVp), B (mAs), C (CT pitch), D (FOV), E (iDose), and F (reconstruction filter). Since each factor could have two or three levels, eighteen groups of factor combinations were organized according to Taguchi’s dynamic algorithm. Three welltrained radiologists ranked the CTA scan images three times for three different phantoms. Thus, 27 (3 × 3 × 3) ranked scores were summed and averaged to imply the integrated performance of one specific group, and eventually, 18 groups of CTA scan images were analyzed. The unique signal-to-noise ratio (S/N, dB) and sensitivity in the dynamic algorithm were calculated to reveal the true contribution of assigned factors and clarify the situation in routine CTA diagnosis. RESULTS: Minimizing the cross-interactions among factors, the optimal factor combination was found to be as follows: A (100 kVp), B (600 mAs), C (pitch 0.200 mm), D (FOV 280 mm), E (iDose 5), and F (filter XCA). The respective MDD values were 2.15, 2.32, and 1.87 mm for 50, 70, and 90 kg phantoms, respectively. The MDD of the 90 kg phantom had the most precise spatial resolution, while that of the 70 kg phantom was the worst. CONCLUSION: The Taguchi static and dynamic optimization algorithms were compared, and the latter’s superiority was substantiated.

This study was aimed to predict the overall survival of colon cancer patients at various (0–IV) stages and provided a robust assessment of the expediency of a surgical operation for such patients. The proposed prediction algorithm was based on the well-known hit and target model adopted for analyzing the cell death from the microscopic viewpoint and implied the application of the Taylor series expansion to the population-based survey dataset (in particular, to the population-based study of colon cancer patients in Taiwan covering the period from 2007 to 2016). In the proposed algorithm, the fundamental degradation of patient's health was represented by a specific function comprising a single exponential term exp(–αt), which was multiplied by an additional term P(αt) that specified the recovery effect of a particular therapy. The revised algorithm successfully predicted the colon cancer patients' overall survival at stages 0–IV and evaluated the expediency necessity of surgical operation for patients at various stages as well. For the above population-based survey dataset, the calculated lethal frequency and average residual life of colon cancer patients who undergone surgical operation amounted to {0.029, 0.036, 0.058, 0.077, 0.236} yr–1 and {34.5, 27.8, 17.2, 13.0, 4.2} yr for stages 0–IV, respectively, while those for the respective groups of patients with no surgical operation was assessed as {0.116, 0.181, 0.256, 0.203, 0.504} yr–1 and {8.6, 5.5, 3.9, 4.9, 2.0} yr, respectively. The proposed algorithm was also applied to interpret the overall survival of lung cancer patients at stages III and IV and exhibited a partial agreement with the actual collected data.

A five-compartmental biokinetic model of I-131 radioiodine based on in-vivo gamma camera scanning results was developed and successfully applied to nine thyroid cancer patients who were administered 1,110 MBq I-131 in capsules for the residual thyroid gland ablation. The I-131 solution activity among internal organs was analyzed via the revised biokinetic model of iodine recommended by the ICRP-30 and -56 reports. Accordingly, a five-compartmental (stomach, body fluid, thyroid, whole body, and excretion) model was established to simulate the metabolic mechanism of I-131 in thyroid cancer patients, whereas the respective four simultaneous differential equations were solved via a self-developed program run in MATLAB. This made it possible to provide a close correlation between MATLAB simulation results and empirical data. The latter data were collected through in-vivo gamma camera scans of nine patients obtained after 1, 4, 24, 48, 72, and 168 hours after radioactive I-131 administration. The average biological half-life values for the stomach, body fluid, thyroid, and whole body of thyroid cancer patients under study were 0.54±0.32, 12.6±1.8, 42.8±5.1, and 12.6±1.8 h, respectively. The corresponding branching ratios I12, I23, I25, I34, I42, and I45 as denoted in the biokinetic model of iodine were 1.0, 0.21±0.14, 0.79±0.14, 1.0, 0.1, and 0.9, respectively. The average values of the AT dimensionless index used to verify the agreement between empirical and numerical simulation results were 0.056±0.017, 0.017±0.014, 0.044±0.023, and 0.045±0.009 for the stomach, thyroid, body fluid + whole body, and total, respectively. The results obtained were considered quite instrumental in the elucidation of metabolic mechanisms in the human body, particularly in thyroid cancer patients.

Objective: The minimum detectable difference (MDD) of computed tomography (CT) scanned images was quantified and optimized according to an indigenous hepatic phantom, line group gauge and Taguchi [Formula: see text] optimization analysis in this work. Methods: Optimal combinations of CT scan factors in every group with the level organization were judged using the Taguchi analysis, in which every factor was organized into only 18 groups, creating evaluated outcomes with the same confidence as if every factor was analyzed independently. The five practical factors of the CT scan were (1) kVp, (2) mAs, (3) pitch increment, (4) field of view (FOV) and (5) rotation time for one loop of CT scan. Insofar as each factor had two or three levels, the total number of 162 (i.e., [Formula: see text]) combinations was considered. Results: The optimal setting was 120[Formula: see text]kVp, 300[Formula: see text]mAs, 0.641 pitch, 320[Formula: see text]mm FOV and 1.0[Formula: see text]s of rotation time of CT scan. The minimal MDD was 2.65[Formula: see text]mm under 0.39[Formula: see text]mm of the slit depth from the revised Student’s [Formula: see text]-test with a 95% confidence level. In contrast, the MDD of conventional and the best one (no. 7) among all original 18 groups were 3.27[Formula: see text]mm and 2.93[Formula: see text]mm for 0.43[Formula: see text]mm and 0.41[Formula: see text]mm slit depths, respectively. Conclusion: The Taguchi analysis was found very lucrative for the design of imaging analysis in practical diagnosis. The indigenous line group gauge and hepatic phantom also proved to be suitable in simulating the human body in real hepatic carcinoma examination.

In this study, a projection of effective blood concentration (EBC) readings of digoxin is made using the inverse problem algorithm based on clinical data for patients with heart failure diseases. Seven factors, including body surface area (BSA), blood urine nitrogen (BUN), creatinine, sodium (Na), potassium (K), magnesium (Mg) ion readings, and mean arterial pressure (MAP) were compiled with nonlinear regression fit to develop a projection function having 29 terms obtained from an inverse problem algorithm via the default function run in STATISTICA. Accordingly, data collected from the clinical 168 heart failure patients were normalized to be included in same domain range ([Formula: see text]1 to +1), and then calculated by the specific algorithm to optimize the numerical solution to evaluate EBC readings of digoxin. The evaluated first-order regression fit owned an optimal loss function ([Formula: see text]) coupled with correlation coefficient [Formula: see text] = 0.892 and variance of 89.20%. Furthermore, 45 patients having similar clinical syndromes were also adopted to verify the projection and implied with high agreement. The BUN factor dominated the projection and defined as the most significant coefficient in the analysis, and K ion, MAP, BSA, and Mg ion factors exhibited minor contributions to the projection. The repeated trials to lower number of factors from seven to a smaller number (namely 6, 5, 4, 3, 2, and 1) for simplifying method but resulting with unaccepted outcomes, with high loss function values and low linearity. However, the algorithm held its accuracy to handle the verified data that were out of the original bounds. The proposed algorithm demonstrated a useful analysis to handle the drug administration in pharmaceutical field.

Objective: To optimize the minimum detectable difference (MDD) of a cardiac X-ray imaging system using the Taguchi L8(27) analysis and a precise line pair (LP) gauge. Methods: The optimal combination of the four critical factors of the cardiac X-ray imaging system, namely X-ray focus, kilovoltage (kVp), milliamper-seconds (mAs) and source image distance (SID), providing the MDD was calculated via the Taguchi analysis and experimentally verified. Two (low and high) levels were assigned for each factor, and eight combinations of four factors were used to acquire instant X-ray images using an NDT commercial LP gauge (with a gauge length of 64[Formula: see text]mm and a width of [Formula: see text][Formula: see text]mm). The latter had five lines and was split gradually from top to bottom for the inspection of X-ray images, whose quality was ranked by three well-trained radiologists according to the double-blind criterion. The ranking grade was given by sharp contrast, low noise and precision to distinguish the LP. Accordingly, the MDD was derived to represent the spatial resolution of instant X-ray images by the revised Student’s [Formula: see text]-test analysis. The optimal combination of factors was experimentally identified and clinically verified in the follow-up inspections. Results: For the conventional setting, Group No. 7 (which obtained the highest grade among eight groups) and the optimal setting, the obtained MDD values were [Formula: see text], [Formula: see text] and [Formula: see text][Formula: see text]mm, respectively, while the LP (line pair/mm) interpolated from the gauge scale amounted to [Formula: see text], [Formula: see text] and [Formula: see text][Formula: see text]LP/mm, respectively. Conclusion: The Taguchi L8 analysis was proved to be instrumental in optimizing the cardiac X-ray imaging system MDD and is recommended to be used jointly with the revised Student’s [Formula: see text]-test analysis for improving the spatial resolution of instant X-ray images.

The occupational X-rays doses of medical staff in a cardiac catheterization laboratory were evaluated.Four customized acrylic phantoms were used to simulate a patient, medical doctor, assistant, and radiologist to evaluate the in-situ X-rays exposure dose using semiconductor dosimeters. The exposure dose was measured under three scenarios that were preset to imply: no shielding, moderate shielding and complete shielding for the medical staff in the laboratory. The doses were applied by changing the dose area product (DAP) from 11,000 to 500,000mGy·cm(2) in 14 increments.The estimated annual occupational doses for doctors, assistants and radiologists in scenarios I, II, and III were: I) 35.03, 7.78, 1.95; II) 1.95, 0.78, 0.06; and III) 0.19, 0.10, 0.05cSv, respectively. The derived linear regression line of the exposure dose with respect to the DAP were extrapolated to obtain the minimum detectable level (MDL) of DAP for triggering the staff dosimeters. Accordingly, the minimum annual dose was estimated as 0.05cSv. Additional shielding provided measurable protection to the staff. The protective clothing used in scenarios II and III can reduce the original dose from scenario I to ∼3% (scenario II) and ∼0.5% (scenario III). The annual occupational dose also changed with the various X-rays energy settings. The annual dose increased to 126% when the preset X-rays energy was changed from 70 to 100kVp.The semiconductor dosimeter proved to be an adequate tool for measuring low doses and low dose rates under these circumstances. The dose can be reduce of I) 35.03, 7.78, 1.95; to II) 1.95, 0.78, 0.06 (∼3%); or III) 0.19, 0.10, 0.05 (∼0.5%)cSv, respectively according to different protective scenarios.