Determination of CT-to-density conversion relationship for image-based treatment planning systems.

The implementation of tissue inhomogeneity correction in image-based treatment planning will improve the accuracy of radiation dose calculations for patients undergoing external-beam radiotherapy. Before the tissue inhomogeneity correction can be applied, the relationship between the computed tomography (CT) value and density must be established. This tissue characterization relationship allows the conversion of CT value in each voxel of the CT images into density for use in the dose calculations. This paper describes the proper procedure of establishing the CT value to density conversion relationship. A tissue characterization phantom with 17 inserts made of different materials was scanned using a GE Lightspeed Plus CT scanner (120 kVp). These images were then downloaded into the Eclipse and Pinnacle treatment planning systems. At the treatment planning workstation, the axial images were retrieved to determine the CT value of the inserts. A region of interest was drawn on the central portion of the insert and the mean CT value and its standard deviation were determined. The mean CT value was plotted against the density of the tissue inserts and fitted with bilinear equations. A new set of CT values vs. densities was generated from the bilinear equations and then entered into the treatment planning systems. The need to obtain CT values through the treatment planning system is very clear. The 2 treatment planning systems use different CT value ranges, one from -1024 to 3071 and the other from 0 to 4096. If the range is correct, it would result in inappropriate use of the conversion curve. In addition to the difference in the range of CT values, one treatment planning system uses physical density, while the other uses relative electron density.     

Machine learning to parse breast pathology reports in Chinese

Introduction

Large structured databases of pathology findings are valuable in deriving new clinical insights. However, they are labor intensive to create and generally require manual annotation. There has been some work in the bioinformatics community to support automating this work via machine learning in English. Our contribution is to provide an automated approach to construct such structured databases in Chinese, and to set the stage for extraction from other languages.

Methods

We collected 2104 de-identified Chinese benign and malignant breast pathology reports from Hunan Cancer Hospital. Physicians with native Chinese proficiency reviewed the reports and annotated a variety of binary and numerical pathologic entities. After excluding 78 cases with a bilateral lesion in the same report, 1216 cases were used as a training set for the algorithm, which was then refined by 405 development cases. The Natural language processing algorithm was tested by using the remaining 405 cases to evaluate the machine learning outcome. The model was used to extract 13 binary entities and 8 numerical entities.

Results

When compared to physicians with native Chinese proficiency, the model showed a per-entity accuracy from 91 to 100% for all common diagnoses on the test set. The overall accuracy of binary entities was 98% and of numerical entities was 95%. In a per-report evaluation for binary entities with more than 100 training cases, 85% of all the testing reports were completely correct and 11% had an error in 1 out of 22 entities.

Conclusion

We have demonstrated that Chinese breast pathology reports can be automatically parsed into structured data using standard machine learning approaches. The results of our study demonstrate that techniques effective in parsing English reports can be scaled to other languages.

     

Reduction of Isoagglutinin in Intravenous Immunoglobulin (IVIG) Using Blood Group A- and B-Specific Immunoaffinity Chromatography: Industry-Scale Assessment

Background

Hemolysis, a rare but potentially serious complication of intravenous immunoglobulin (IVIG) therapy, is associated with the presence of antibodies to blood groups A and B (isoagglutinins) in the IVIG product. An immunoaffinity chromatography (IAC) step in the production process could decrease isoagglutinin levels in IVIG.

Objectives

Our objectives were to compare isoagglutinin levels in a large number of IVIG (Privigen^®) batches produced with or without IAC and to assess the feasibility of the production process with an IAC step on an industrial scale.

Methods

The IAC column comprised a blend of anti-A and anti-B resins formed by coupling synthetic blood group antigens (A/B-trisaccharides) to a base bead matrix, and was introduced towards the end of the industrial-scale IVIG manufacturing process. Isoagglutinin levels in IVIG were determined by anti-A and anti-B hemagglutinin direct and indirect methods according to the European Pharmacopoeia (Ph. Eur.) and an isoagglutinin flow cytometry assay. IVIG product quality was assessed with respect to the retention of immunoglobulin G (IgG) subclasses, specific antibodies, and removal of IgM using standardized procedures.

Results

The IAC step reduced isoagglutinins in IVIG by two to three titer steps compared with lots produced without IAC. The median anti-A and anti-B titers with IAC were 1:8 and 1:4, respectively, when measured by the Ph. Eur. direct method, and 1:2 and <1, respectively, when measured by the Ph. Eur. indirect method. The isoagglutinin flow cytometry assay showed an 87–90 % reduction in isoagglutinins in post-IAC versus pre-IAC fractions. IAC alone reduced anti-A and anti-B of the IgMs isotype by 92.5–97.8 % and 95.4–99.2 %, respectively. Other product quality characteristics were similar with and without IAC.

Conclusions

IAC is an effective method for reducing isoagglutinin levels in IVIG, and it is feasible on an industrial scale.

     

Clinical course of neuropathologically confirmed frontal-variant Alzheimer's disease

Background A 66-year-old man presented with a 3-year history of personality changes marked by increasing apathy, social withdrawal and deficits in complex attention, and a 1-year history of progressive memory problems and difficulties in planning and carrying out complex tasks. Investigations Three neuropsychological examinations over 2 years, neurological examination, routine laboratory tests, brain MRI, single-photon emission CT scan, genetic analyses, and neuropathological examination. Diagnosis A clinical diagnosis of frontal-variant frontotemporal dementia was superseded by postmortem neuropathological evidence, which established a diagnosis of frontal-variant Alzheimer's disease. Management The patient and his spouse were referred for counseling, and the patient was referred for follow-up examinations.     

Expression of glia maturation factor in neuropathological lesions of Alzheimer's disease

R. Thangavel, D. Stolmeier, X. Yang, P. Anantharam and A. Zaheer (2012) Neuropathology and Applied Neurobiology 38, 572–581 Expression of glia maturation factor in neuropathological lesions of Alzheimer's disease Aims: The pathology of Alzheimer's disease (AD) is characterized by the presence of amyloid plaques (APs), neurofibrillary tangles (NFTs), degenerating neurones, and an abundance of reactive astrocytes and microglia. We aim to examine the association between glia maturation factor (GMF) expression, activated astrocytes/microglia, APs and NFTs in AD‐affected brain regions. Methods: Brain sections were stained with Thioflavin‐S to study AD pathology and sequentially immunolabeled with antibodies against GMF, glial fibrillary acidic protein (marker for reactive astrocytes), and Ionized calcium binding adaptor molecule 1 (Iba‐1, marker for activated microglia) followed by visualization with avidin‐biotin peroxidase complex. Results: Our double immunofluorescence labelling with cell‐specific markers demonstrated the glial localization of GMF. The immunohistochemical data showed that APs and NFTs are associated with increased expression of GMF in reactive glia of AD brains compared with non‐AD controls. Conclusions: This is the first report that shows GMF, a mediator of central nervous system inflammation, is expressed in the brain regions affected in AD and that GMF is mainly localized in reactive astrocytes surrounding APs/NFTs. The distribution of GMF‐immunoreactive cells in and around Thioflavin‐S stained APs and NFTs suggests involvement of GMF in inflammatory responses through reactive glia and a role of GMF in AD pathology.