| 1. |
- Bornefalk, Hans
(författare)
-
Estimation and Comparison of CAD System Performance in Clinical Settings
- 2005
-
Ingår i: Academic Radiology. - : Elsevier BV. - 1076-6332 .- 1878-4046. ; 12:6, s. 687-694
-
Tidskriftsartikel (refereegranskat)abstract
- Rationale and Objectives. Computer-aided detection (CAD) systems are frequently compared using free-response receiver operating characteristic (FROC) curves. While there are ample statistical methods for comparing FROC curves, when one is interested in comparing the outcomes of 2 CAD systems applied in a typical clinical setting, there is the additional matter of correctly determining the system operating point. This article shows how the effect of the sampling error on determining the correct CAD operating point can be captured. By incorporating this uncertainty, a method is presented that allows estimation of the probability with which a particular CAD system performs better than another on unseen data in a clinical setting. Materials and Methods. The distribution of possible clinical outcomes from 2 artificial CAD systems with different FROC curves is examined. The sampling error is captured by the distribution of possible system thresholds of the classifying machine that yields a specified sensitivity. After introducing a measure of superiority, the probability of one system being superior to the other can be determined. Results. It is shown that for 2 typical mammography CAD systems, each trained on independent representative datasets of 100 cases, the FROC curves must be separated by 0.20 false positives per image in order to conclude that there is a 90% probability that one is better than the other in a clinical setting. Also, there is no apparent gain in increasing the size of the training set beyond 100 cases. Discussion. CAD systems for mammography are modeled for illustrative purposes, but the method presented is applicable to any computer-aided detection system evaluated with FROC curves. The presented method is designed to construct confidence intervals around possible clinical outcomes and to assess the importance of training set size and separation between FROC curves of systems trained on different datasets.
|
|
| 2. |
- Chai, Chun-Ming, et al.
(författare)
-
Signs in Vector-Electrocardiography (VECG) Predicting the Fibrillatory Propensity of Iodixanol and Mannitol Solutions After Injection Into the Left Coronary Artery of Pigs.
- 2007
-
Ingår i: Academic Radiology. - : Elsevier BV. - 1878-4046 .- 1076-6332. ; 14:2, s. 162-177
-
Tidskriftsartikel (refereegranskat)abstract
- Rationale and Objectives. To find signs in vector-electrocardiography (VECG) predicting the ventricular fibrillatory propensity (VF-PROP) of iodixanol and mannitol solutions after injection into the left coronary artery (LCA) of pigs. Materials and Methods. Five plasma-isotonic solutions perfused LCA: lod 320 + Na/Ca (iodixanol 320 mg I/mL, 19 mM NaCl, 0.3 mM CaCl2), lod 320 + Mann (iodixanol 320 mg I/mL, 50 mM mannitol), Mann + Na/Ca (240 mM mannitol, 19 mM NaCl, 0.3 mM CaCl2) Mann (275 mM mannitol), and Ringer (representing "physiologic electrolytes"). The first two solutions have at 37 degrees C viscosity 13 mPas and the others <1 mPas. In eight pigs, 20 mL of each solution was injected twice for 10 seconds, and in 15 pigs, each solution was injected for 11-40 seconds (0.5 mL/second) through a wedged catheter in the LCA. If ventricular fibrillation (VF) occurred, injection was stopped and heart was defibrillated. If VF did not occur, perfusion period was 40 seconds. A higher frequency of VF and a shorter period from start of injection until start of VF gave a solution a higher ranking of VF-PROP. Results. The 10-second injections caused no VF. Ringer and lod 320 + Na/Ca caused no VF after 40-second injections, whereas the other solutions caused VF. Ranking the solutions from lowest to highest VF-PROP gave: Ringer = lod 320 + Na/Ca < lod 320 + Mann < Mann + Na/Ca < Mann. Prolongation of QRS time and QTc time were the only VECG signs that showed significant differences (P <.05) between all solutions and correctly ranked the VF-PROP of all solutions in both animal groups. Conclusion. The results fit with the concept that a more physiologic electrolyte composition and a higher viscosity of a test solution will, after start of injection of that solution into LCA, delay changes in the electrolyte composition in myocardial interstitial fluid and also delay start of VF. If a plasma isotonic contrast medium (CM) with lower viscosity than that of iodixanol at 320 mgI/mL were created, we conclude that such a CM should have electrolyte composition closer to that of Ringer than present composition (19 mM NaCl and 0.3 mM CaCl2) to counteract the effects of faster diffusion of nonphysiologic electrolyte composition from the low-viscosity CM to myocardial interstitial fluid.
|
|
| 3. |
- Dapi Nzefa, Léonie, 1971-, et al.
(författare)
-
"I eat to be happy, to be strong, and to live." perceptions of rural and urban adolescents in Cameroon, Africa.
- 2007
-
Ingår i: Journal of nutrition education and behavior. - US : Elsevier. - 1499-4046 .- 1878-2620. ; 39:6, s. 320-326
-
Tidskriftsartikel (refereegranskat)abstract
- OBJECTIVE: To investigate factors influencing rural and urban adolescents' food perceptions during a time of nutritional transition in Cameroon, Africa.DESIGN: Qualitative in-depth interviews.SETTINGS: Yaoundé urban and Bandja rural areas.PARTICIPANTS: Fifteen adolescents 12 to 15 years old purposely selected from schools in urban and rural areas.ANALYSIS: Interviews were audiotaped, transcribed, and analyzed using Grounded Theory method.FINDINGS: Factors influencing adolescents' food perceptions from the rural area were "to live" "health" and "poverty." Among adolescents from the urban poor area, "health," "beauty," and "not enough money" were factors. Among adolescents from the urban rich area, "pleasure" and "beauty" were factors. Rural girls liked "to be fat," whereas girls from the urban poor wanted "to be a little bit fat," and girls from the urban rich wanted "to be normal."CONCLUSIONS AND IMPLICATIONS: Food behavior is changing from a diet composed of traditional food in rural areas to a more westernized diet in urban areas. The relationship between socioeconomic factors and nutrition needs to be examined with a sufficiently large number of adolescents to investigate these factors in a quantitative survey. Healthful local food should be available at home and from vendors. Nutrition education about food and diet-related diseases is needed in school.
|
|
| 4. |
- Diaz, Sandra, et al.
(författare)
-
Progression of Emphysema in a 12-month Hyperpolarized (3)He-MRI Study Lacunarity Analysis Provided a More Sensitive Measure than Standard ADC Analysis(1).
- 2009
-
Ingår i: Academic Radiology. - : Elsevier BV. - 1878-4046 .- 1076-6332. ; 16:6, s. 700-707
-
Tidskriftsartikel (refereegranskat)abstract
- RATIONALE AND OBJECTIVES: Inhaled hyperpolarized (3)He magnetic resonance (MR) imaging has been used to measure alveolar size in patients with emphysema. The aim of this study was to test the hypothesis that (3)He MR images could be used to develop a biomarker of emphysema progression. MATERIALS AND METHODS: Twelve healthy controls and 18 patients with emphysema (eight current smokers, 10 ex-smokers) were imaged at baseline and 6 and 12 months. An additional nine subjects with alpha-1 antitrypsin deficiency (four with emphysema, six without symptoms) were also imaged at baseline and at 6 months. Each subject was imaged at two lung volumes: functional residual capacity (FRC) and FRC plus 15% of total lung capacity. Means and standard deviations of apparent diffusion coefficients (ADCs) were calculated from coronal images of the entire lung and correlated with pulmonary function test results. The lacunarity hypothesis was tested and calculated from the data using a range of 2x2 x 2 to 6x6 x 6 voxels, and the average was calculated. RESULTS: There was no change in the mean ADC at either lung volume in any subject over the 6- or 12-month period. FRC and residual volume increased over the 12 months, suggesting air trapping. The lacunarity of images collected at FRC increased at 6 and 12 months in smokers only (P=.063 and P=.023, respectively). CONCLUSIONS: The mean ADC calculated from MR images of the lungs with helium was not sufficiently sensitive to detect changes over a 12-month period. However, lacunarity captured more of the spatial information in the images and detected emphysema progress in the smokers.
|
|
| 5. |
|
|
| 6. |
- Lidegran, Marika, et al.
(författare)
-
Cranial CT for diagnosis of intracranial complications in adult and pediatric patients during ECMO : Clinical benefits in diagnosis and treatment
- 2007
-
Ingår i: Academic Radiology. - Netherlands : Elsevier BV. - 1076-6332 .- 1878-4046. ; 14:1, s. 62-71
-
Tidskriftsartikel (refereegranskat)abstract
- RATIONALE AND OBJECTIVES: To evaluate the clinical utility of cranial computed tomography (CT) in pediatric and adult patients during ongoing extracorporeal membrane oxygenation (ECMO) treatment from acute respiratory failure and to assess the frequency of intracranial hemorrhage (ICH) and infarction during the treatment. MATERIALS AND METHODS: The medical records of 123 consecutive patients, 54 children (ages 3 months-17 years) and 69 adults (ages 18-62 years), treated with ECMO over a 10-year period were searched for cranial CT performed during ECMO. Indications for CT, CT findings, impact on clinical management, and patient outcome were noted. In addition, all CT scans were reviewed for the frequency of ICH or infarction. RESULTS: Seventy-eight patients had cranial CT while on ECMO. ICH or cerebral infarction were detected in 45 (37%) of the 123 patients. Eighteen patients (15%) had focal hemorrhage, 11 (9%) focal infarction, and 16 (13%) general brain edema. In 16 of the 45 patients, the CT findings were decisive to withdraw the ECMO treatment. Five patients were weaned from ECMO, and in four patients the findings motivated cranial surgery during ECMO. In the remaining 20 patients with less extended intracranial pathology, the ECMO treatment was continued with high survival. CONCLUSION: Cranial CT has an important role during ECMO treatment to reveal or exclude severe intracranial complications where ECMO treatment should be discontinued. Less severe complications have a favorable prognosis with continued treatment. Our study suggests an underreporting of intracranial complications in adults and pediatric patients on ECMO because of low utilization of neuroimaging.
|
|
| 7. |
|
|
| 8. |
|
|
| 9. |
- Zakhary, Mina M., et al.
(författare)
-
Prevalence and Etiology of Intracranial Hemorrhage in Term Children Under the Age of Two Years: A Retrospective Study of Computerized Tomographic Imaging and Clinical Outcome in 798 Children
- 2009
-
Ingår i: Academic Radiology. - : Elsevier BV. - 1878-4046 .- 1076-6332. ; 16:5, s. 572-577
-
Tidskriftsartikel (refereegranskat)abstract
- Rationale and Objectives. The purposes of this study were to retrospectively identify various etiologies underlying intracranial hemorrhages (ICHs) in term infants aged <2 years and their respective prevalence in this population and to describe the long-term clinical outcomes in these patients. Materials and Methods. A retrospective review of the medical records and computed tomographic studies of the head in 798 term infants aged 0 to 24 months with suspected or known ICHs was conducted. Results. ICHs were present in 195 of the 798 infants (24%). More than one type of ICH was present in 32%. Subdural hemorrhage was the most frequent type of ICH, occurring in 63% of the infants. Good clinical outcomes were present in 49% of the infants but varied depending on the location, etiology, and timing of the ICH. Conclusion. The incidence of various etiologies of ICH depended on the ages of the infants. The overall clinical outcomes were good, with no long-term sequelae in half of the infants presenting with ICHs. In infants aged >4 weeks presenting with ICHs, special attention should be given to the possibility of nonaccidental trauma etiology, because this is common and has worse long-term outcomes.
|
|
