Principal component analysis of dynamic positron emission tomography images

Multivariate image analysis can be used to analyse multivariate medical images. The purpose could be to visualize or classify structures in the image. One common multivariate image analysis technique which can be used for visualization purposes is principal component analysis (PCA). The present work concerns visualization of organs and structures with different kinetics in a dynamic sequence utilizing PCA. When applying PCA on positron emission tomography (PET) images, the result is initially not satisfactory. It is illustrated that one major explanation for the behaviour of PCA when applied to PET images is that it is a data-driven technique which cannot separate signals from high noise levels. With a better understanding of the PCA, gained with a strategy of examining the image data set, the transformations, and the results using visualization tools, a surprisingly easily understood methodology can be derived. The proposed methodology can enhance clinically interesting information in a dynamic PET imaging sequence in the first few principal component images and thus should be able to aid in the identification of structures for further analysis.

Grading of transitional cell bladder carcinoma by texture analysis of histological sections

Conventional subjective visual grading of malignancy in bladder carcinoma suffers from poor standardization. In an attempt to improve this by image analysis we carried out a texture analysis of a total of 197 images from Feulgen-stained histological tissue sections. The textural features used were based on the spatial gray-tone co-occurrence probability matrices for step-lengths 3, 4 and 6 pixels. These features were compared to the subjective grading by multivariate statistical methods. Using a two-stage hierarchical classifier 84.3% agreement was obtained between the subjective and computer-based classification. This was verified through the jackknifed classification technique.

PACS for radiology conferences--improvement of application software

The introduction of PACS in a radiology department means that most functions that are available in the film-based system must be included. One important function is the radiology conference. The handling and application programs of digital workstations are normally not developed for demonstrations of many images in a limited time. This paper describes a workstation with specially designed software for radiology conferences. The application is separated in preparation and presentation of the cases to be demonstrated. Very fast image handling is achieved during the conference because the function is based on the principle of prefetching of the selected images. The images are presented on a large screen with high resolution. The experiences of digital radiology conferences are good. However, reference films from previous examinations create extra work because they have to be shown on the conventional lightbox.

MUSE--a new tool for interactive image analysis and segmentation based on multivariate statistics

MUSE--a new software tool for the interactive exploration of multivariate images and the development of image segmentation methods has been designed, implemented, and tested in a number of real application projects. The multivariate statistical classification and projection methods in MUSE can be used not only to analyze multispectral images but also, in special cases, multitemporal images and volume images. Additionally MUSE can be applied to normal greyscale images provided they are made multivariate through an initial processing step. This step may consist in the application of filters designed to enhance any existing texture differences between different regions in the images. MUSE has been successfully applied to medical images (color photographs, MR, PET, SPECT) as well as to satellite images (Landsat TM).

Differences in side effects between a conventional carbamazepine preparation and a slow-release preparation of carbamazepine

The aim of this double-blind cross-over study was to investigate whether side effects of carbamazepine (CBZ) could be reduced by using a slow-release CBZ preparation. Twenty-one adult patients with epilepsy who had side effects related to the use of CBZ took part in the trial. Patients were randomized to receive either a conventional (C) or slow-release (SR) CBZ preparation for 3 months and were then switched over to the other preparation for another 3 months. The daily dose and dosing frequency of CBZ were kept the same as before the study. The quality and severity of side effects were assessed monthly using a scored questionnaire containing questions about systemic toxicity (STRS) and neurotoxicity (NTRS). Twenty patients could be evaluated. The mean total values of NTRS of 3 monthly visits on each drug were significantly less during SR than during C treatment (P less than 0.05). All the items of NTRS scored lower during SR therapy, and the difference was significant for the occurrence of headache, dizziness and disturbances of vision, speech and coordination. The total score of STRS was also lower during SR, but the difference was not significant. Eleven patients preferred SR, 3 preferred C and 6 patients estimated the periods to be equal. In conclusion, a slow-release preparation of CBZ can render fewer side effects than conventional CBZ preparations.

The measuring of cell features

The usefulness of various feature sets for discriminating between different cell populations cannot be assessed without considering the entire automated cytology system, from specimen preparation through scanning, cell search, cell segmentation, artifact rejection and feature extraction to object classification and multivariate data analysis methods. These different stages in the analysis procedure are discussed in this paper and illustrated by examples from the author's experience of cell analysis studies, ranging from single-cell classification experiments to a test of a complete automated system under realistic conditions on a set of 397 cell specimens.

Image cytometry in malignancy grading of breast cancer. Results in a prospective study with seven years of follow-up

A fully automated system for image cytometry is presented. In combination with a special tumor sampling procedure, the system proved able to record with high reproducibility a large number of tumor cell population descriptors of high prognostic significance in a prospective study of breast cancer. These included the degree of abnormality of DNA distribution (strongly correlated with the proliferative rate) and some descriptors of shape and texture (strongly correlated with the grading system for breast cancer suggested by the World Health Organization). While the extreme complexity of a true malignancy grading system for breast cancer is underlined, the results of this study indicate that its creation is a possibility, given sufficient effort by interdisciplinary teams using modern sophisticated analytical techniques.

Monthly antimalarial chemotherapy to children in a holoendemic area of Liberia

Two hundred and eighty-two children, two to nine years old, were included in a prospective three-year study in four villages with holoendemic malaria. In three villages the children received monthly doses of either chloroquine, pyrimethamine or chlorproguanil respectively for two years. In the fourth, vitamin tablets were used as placebo. Presumptive treatment with chloroquine (10 mg base kg-1) was given to all children with fever of suspected malarial origin. The two-year drug distribution was satisfactorily fulfilled to 168 children. Surveys, including physical and laboratory examinations were performed every six months, four weeks after medication. A fifth village was only visited at the start of the study and after two years. The mean crude parasite rate was initially 92%. Plasmodium falciparum was the main species. Splenomegaly was recorded in all children. In the chloroquine-treated children, the parasite rates varied between 30% and 50% during the study. By the end of the second year the spleen rate was reduced from 100% to 50%. Reported episodes of fever were reduced to half and mean haematocrit levels increased by 6% in comparison with children receiving the placebo. Total IgG concentrations were reduced from 36.7 g l-1 to 25.9 g l-1, whereas no significant decrease was observed in malarial seropositivity as measured by indirect immunofluorescence. Chlorproguanil had a weaker impact on parasitaemia with parasite rates between 50% and 90%. However, the spleen rate was reduced to 67% and there was a significant reduction of reported fever episodes. Mean haematocrits increased by 4%. Total IgG decreased from 31.8 g l-1 to 23.8 g l-1. In contrast, in the pyrimethamine group, the placebo group and the untreated group from the fifth village, the malariometric indices after two years were comparable to each other and to the initial values. During the third year only presumptive chloroquine treatment was given, and by the end of the study all malariometric indices were again comparable. From clinical observations there was no apparent impairment of protective immunity to malaria from the two years of regular distribution of the drugs. We conclude that a certain degree of malaria control could be achieved in Liberian children by the administration of monthly doses of chloroquine 10 mg base kg-1. The administration of chlorproguanil (1.5 mg kg-1) represents an alternative regimen.

In vivo response of Plasmodium falciparum to different doses of chloroquine in semi-immune children in Liberia, West Africa

The efficacy of different doses of chloroquine in suppressing patent parasitaemia was investigated in 326 children two to 12 years old, living in six villages with holoendemic malaria. The children were given single doses (2, 3, 5-7 or 9-12 mg base kg-1) or a standard treatment over three days (25 mg base kg-1). Parasite prevalences were recorded after one, two, three, four, six and eight weeks. Complete clearance of Plasmodium falciparum trophozoites (TC) by day 7 was achieved by a dosage of 9-12 mg kg-1. By probit analysis of log dose response, 50% clearance (TC50) was established at about 1.5 mg kg-1, whereas a TC95 required 5.5 mg kg-1. The reappearance of patent parasitaemia was dependent on the dose of chloroquine given and on malaria transmission. After the standard dose treatment, only one re-infection in 56 children appeared within 21 days despite high sporozoite inoculation rates in the area. The dosage of 9-12 mg kg-1 yielded a hundredfold reduction of mean parasite density in the children if calculated over a four-week period. It may represent a suitable monthly regimen in a malaria control scheme in a holoendemic area with high P. falciparum sensitivity to chloroquine.

Malaria control by chlorproguanil. I. Clinical effects and susceptibility of Plasmodium falciparum in vivo after seven years of monthly chlorproguanil administration to children in a Liberian village

For seven years, chlorproguanil (1.0 to 2.0 mg kg-1) was administered monthly to the children below 15 years of age in a village with holoendemic malaria. Malariometric indices were recorded every six months. Susceptibility in vivo was monitored by the clearance of Plasmodium falciparum parasitaemia after drug intake. Three parasite species were found initially: P. falciparum (52%), P. malariae (8%) and P. ovale (4%). The parasites found during the study were mainly P. falciparum, and parasite rates ranged from 37 to 87% at the different surveys one month after respective drug intake. A fifty-fold decrease of mean parasite density was generally observed seven days after drug intake. Splenomegaly was initially recorded in all two to nine year old children, with a mean size of 2.64 according to Hackett's index. From 18 months onwards as the mean spleen index was 1.15 in the same age group. Chlorproguanil may represent an important alternative drug to groups at risk in malaria control schemes.

Different malaria control activities in an area of Liberia--effects on malariometric parameters

The epidemiology of malaria was studied in a West African mining town (Yekepa) and three surrounding zones defined as Close, Middle and Far areas. Malariometric parameters were investigated in children two to nine years of age at the end of the rainy season. In Yekepa, vector control measures and intense suppression of malaria with drugs had created an almost hypoendemic situation with a spleen rate of 11%. In Close area, vector control was applied to some extent and malaria drugs were frequently used for treatment; the spleen rate was 40%. In Middle area, a mobile clinic provided sporadic malaria treatment to small children, but the clinic did not reach out to Far area. The spleen rates were 95 and 99%, respectively. Three species of Plasmodium were found in all areas. The prevalences in Far area were P. falciparum 82%, P. malariae 39% and P. ovale 9%. The crude parasite rates increased from 13% in Yekepa to 92% in Far area, whereas haematocrit levels decreased from 37.6 to 35.2, respectively. Plasmodium falciparum seropositivity, as measured by indirect immunofluorescence, was 74% in Yekepa and 99% in Middle and Far areas. Total IgG concentrations ranged from 18 g1(-1) in Yekepa to 33 g1(-1) in Far area. Three main anopheline species were found in the zones outside Yekepa. Their relative frequencies in Far area were Anopheles funestus 45%, A. hancocki 37%, and A. gambiae 18%. The local inoculation rates gradually increased outwards from Yekepa from less than 0.01 to 0.17 inoculations per man and night at the beginning of the dry season.(ABSTRACT TRUNCATED AT 250 WORDS)

Rift Valley fever virus vaccine trial: study of neutralizing antibody response in humans

The serological response to immunization with a formalin inactivated Rift Valley fever (RVF) vaccine was studied in 963 Swedish UN soldiers serving in the Sinai peninsula. Antibody titres were determined with a plaque reduction neutralization test (PRNT). Attempts were made to give all soldiers three injections (1 ml s.c. days 0, 7-10 and 28-30) but 128 soldiers received only two injections. In a group of 51 fully vaccinated individuals, repeated blood samples were collected. Fifty of the vaccinees seroconverted. Serum collected six weeks after the first vaccination revealed the highest antibody titres. The geometric mean titre then decreased rapidly during the following two weeks. Six months after vaccination sera were collected from 433 vaccinees who had received three injections and 379 had antibodies detectable by PRNT (88% PRNT greater than or equal to 10). The corresponding figures one and two years after vaccination were 223 seropositives out of 255 (91% PRNT greater than or equal to 10) and 91 out of 123 (74% PRNT greater than or equal to 10), respectively. Multiple stepwise regression showed that three injections gave a better antibody response than two injections. This analysis also showed that the magnitude of the antibody response was reduced with increasing age. Slight, local and general side effects were reported in 6% of the vaccinees and these reactions occurred in individuals with relatively higher antibody response.

A program for logistic prediction modelling

A computer program has been developed that can be used for analysing a binary outcome variable and a set of regressors of type interval with a logistic (i.e. nonlinear) model.

Is the working capacity of Liberian industrial workers increased by regular malaria prophylaxis?

In a study of the impact of malaria prophylaxis upon the physical working capacity of Liberian industrial workers, two groups of men, one with and the other without malaria prophylaxis, were compared over a period of one year. At the beginning and at the end of the study, the haemoglobin concentration, haematocrit, blood volume and physical performance--measured by bicycle ergometry and expressed as work load at heart rate 170--were compared. No significant differences were found, either within or between the two groups. Routinely distributed malaria prophylaxis thus seems to be of little importance with respect to working capacity in this type of community, where malaria is meso-endemic.

Treatment of non-invasive amoebiasis--a comparison between tinidazole and metronidazole

In a small pilot study the parasite cure rate (PCR) of non-invasive amoebiasis was compared after treatment with metronidazole 800 mg three times daily or tinidazole 600 mg twice daily for five days. Both treatment regimens were found to be highly unfavourable with PCRs of 44 and nil respectively, in contrast to previous published results showing PCR over 80%.

A long-term follow up study of amoebiasis treated with metronidazole

Metronidazole, in a dose of 40 mg/kg body weight in 3 divided doses for 10 days was used to treat 194 patients infected with Entamoeba histolytica. When 166 subjects were checked 1 month after treatment, with 3 stool samples taken on different days, 20 failures were found giving a parasitic cure rate of 88%. A lower-than-expected failure rate was noted in children, immigrants and asymptomatic subjects. No more failures were found in 78 patients who went through an extended follow-up, with examinations after 4 and 10-12 months.

A case-control study in northern Liberia of Plasmodium falciparum malaria in haemoglobin S and beta-thalassaemia traits

A case-control study was carried out on 558 patients with malaria attending a hospital in Yekepa, northern Liberia; 94 patients (16.8%) were aged at least ten years, probably because of a low level of protective immunity in town dwellers due to malaria control. The proportion of sickle cell traits (1.8%) among the patient group was lower than in the population (7.2%) served by the hospital (chi 2, 21.455, 1 df, P less than 0.001). A stratified analysis showed the relative risk for Plasmodium falciparum malaria in sickle cell trait over normal homozygotes, as 0.29 (upper 95%) confidence interval, 0.56). For beta-thalassaemia trait, the proportion among patients was 5.5% as against 9.0% in the general population (chi 2, 6.158, 1 df, 0.025 greater than P greater than 0.010). Stratified analysis gave a weighted relative risk for beta-thalassaemia heterozygotes of 0.49 (upper 95% confidence interval, 0.74). Although there were four beta-thalassaemia traits in the 10-14 year stratum with moderate to high parasitaemias, we consider that the overall results are consistent with relative resistance against P. falciparum malaria of both sickle cell and beta-thalassaemia heterozygotes in this population. No conclusions were possible from this investigation with regard to HbC and the malaria hypothesis. We found no evidence that P. falciparum malaria elevates HbA2 concentrations into the beta-thalassaemia range.