P–045 A retrospective study on Hyaluronic acid binding sperm selection in ICSI cycles

Study question

Does Hyaluronic acid(HA) binding sperm selection prior ICSI produce better outcome compare to PVP?

Summary answer

No significant difference in fertilization rate, blastocyst utilization rate, pregnancy rate and miscarriage rate in HA vs PVP group.

What is known already

HA is natural, non-sulphated glycosaminoglycan secretion that abundantly found in the cervical mucus and the cumulus oophorus complex (COC). In in-vivo, HA binding sites on the sperm plasma membrane indicate sperm maturity, mature sperm reaching HA-coated COC can bind and digest HA and subsequent hyperactivation further facilitate fertilization. In IVF, the use of HA tries to mimic in-vivo to select mature sperm with high DNA integrity prior to ICSI. In HA, movement of mature sperm is ‘slowed’ thus allow the selection of sperm to be used in ICSI. Sperm immaturity is known to associated with aneuploidy incidence in the embryos.

Study design, size, duration

ICSI cycle using HA (N = 83) was adapted from January to December 2020 while ICSI cycle using PVP (N = 133) was adapted from January 2018 to December 2019. Mean age of patient were 35.64±4.33 vs 34.15±4.75 for HA vs PVP group respectively. Fertilization rate, blastocyst utilization rate, pregnancy rate and miscarriage rate were recorded. This study included all ICSI cycle and both frozen and fresh embryo transfer data. Surgical retrieved sperm was excluded from this study.

Participants/materials, setting, methods

A 1.5ul of treated spermatozoa suspension was connected with a pipette tip to a 5ul droplet of fresh holding medium (Cook Gamete Buffer). Simultaneously, a 5ul droplet of HA medium (Origio SpermSlow) was connected to the 5ulL droplet of holding medium in a ICSI dish (Sparmed Oosafe) covered with oil (Vitrolife Ovoil). Sperm which were ‘slowed’ in the Sperm slow droplets with normal morphology according to WHO 2010 guideline were selected for ICSI at 400X.

Main results and the role of chance

The fertilization rate of the HA vs PVP- binding sperm are 68.6% vs 66.2%. As for the blastocyst utilisation rate is 61.6% vs 73.22% for HA vs PVP- binding sperm group. Pregnancy and miscarriage rate for HA vs PVP are 42.3% vs 51.5% and 19.4% vs 26.2% respectively.

There was no significant difference in the fertilization rates, blastocysts utilisation rate, pregnancy rate and miscarriage rate between the HA vs PVP- binding sperm groups (P < 0.05). However, a trend of higher pregnancy (51.5% vs 42.3%; P = 0.279) and miscarriage rate were observed in the PVP group (26% vs 19%; P = 0.545) as compared to the HA group, but the difference was not statistically significant. The reason behind this might be the HA assist to select the mature sperm with higher DNA integrity and low frequency of chromosomal aneuploidies which contribute to the lower miscarriage rate in the study.

Limitations, reasons for caution

This is a retrospective study on HA binding sperm selection vs PVP prior to ICSI.

Further research which includes a large number of RCT sample size should be warranted.

Wider implications of the findings: The HA- sperm binding selection ICSI might only be beneficial to certain group of patients (high- DNA fragmentation sperm). A larger RCT study may be necessary to establish a relationship between HA-sperm binding selection vs aneuploidy rate via PGT analysis.

Trial registration number

Not applicable

Abstract GS5-04: Prediction of occult invasive disease in ductal carcinoma in situ using deep learning features

Background: Deep learning, especially deep convolutional neural network (CNN), has emerged as a promising approach for many image recognition or classification tasks, demonstrating human or even superhuman performance. Used as feature extractor, some pre-trained CNN models can match or surpass the performance of domain-specific, “handcrafted” features. In this study, we aim to determine whether deep features extracted from digital mammograms using a pre-trained deep CNN are prognostic of occult invasive disease for patients with ductal carcinoma in situ (DCIS) on core needle biopsy.

Materials and Methods: In this retrospective study, we collected digital mammography magnification views for 99 subjects with DCIS at biopsy, 25 of which were subsequently upstaged to invasive cancer. We utilized a deep CNN model that was pre-trained on non-medical images (e.g., animals, plants, instruments) as the feature extractor. Through a statistical pooling strategy, we extracted deep features at different levels of convolutional layers from the lesion areas, without sacrificing the original resolution or distorting the underlying topology. A multivariate classifier was then trained to predict which tumors contain occult invasive disease. This was compared to the performance of traditional “handcrafted” computer vision (CV) features previously developed specifically to assess mammographic calcifications. The generalization performance was assessed using Monte Carlo cross validation and receiver operating characteristic (ROC) curve analysis.

Results: Deep features were able to distinguish DCIS with occult invasion from pure DCIS, with an area under the ROC curve (AUC-ROC) equal to 0.70 (95% CI: 0.68-0.73). This performance was comparable to the "handcrafted" CV features (AUC-ROC = 0.68, 95% CI: 0.66-0.71) that were designed with prior domain knowledge.

Conclusion: In spite of being pre-trained on only non-medical images, the deep features extracted from digital mammograms demonstrated comparable performance to "handcrafted" CV features for the challenging task of predicting DCIS upstaging.

Acknowledgments: This work was supported in part by NIH/NCI R01-CQA185138 and DOD Breast Cancer Research Program W81XWH-14-1-0473.

Citation Format: Lo JY, Grimm LJ, Mazurowski MA, Baker JA, Marks JR, King LM, Maley CC, Hwang E-SS, Shi B. Prediction of occult invasive disease in ductal carcinoma in situ using deep learning features [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr GS5-04.

Bayesian networks of BI-RADStrade mark descriptors for breast lesion classification

We investigated Bayesian network structure learning and probability estimation from mammographic feature data in order to classify breast lesions into different pathological categories. We compared the learned networks to naive Bayes classifiers, which are similar to the expert systems previously investigated for breast lesion classification. The learned network structures reflect the difference in the classification of biopsy outcome and the invasiveness of malignant lesions for breast masses and microcalcifications. The difference between masses and microcalcifications should be taken into consideration when interpreting systems for automatic pathological classification of breast lesions. The difference may also affect use of these systems for tasks such as estimating the sampling error of biopsy.

Computerized classification of suspicious regions in chest radiographs using subregion Hotelling observers

We propose to investigate the use of subregion Hotelling observers (SRHOs) in conjunction with perceptrons for the computerized classification of suspicious regions in chest radiographs for being nodules requiring follow up. Previously, 239 regions of interest (ROIs), each containing a suspicious lesion with proven classification, were collected. We chose to investigate the use of SRHOs as part of a multilayer classifier to determine the presence of a nodule. Each SRHO incorporates information about signal, background, and noise correlation for classification. For this study, 225 separate Hotelling observers were set up in a grid across each ROI. Each separate observer discriminates an 8 by 8 pixel area. A round robin sampling scheme was used to generate the 225 features, where each feature is the output of the individual observers. These features were then rank ordered by the magnitude of the weights of a perceptron. Once rank ordered, subsets of increasing number of features were selected to be used in another perceptron. This perceptron was trained to minimize mean squared error and the output was a continuous variable representing the likelihood of the region being a nodule. Performance was evaluated by receiver operating characteristic (ROC) analysis and reported as the area under the curve (Az). The classifier was optimized by adding additional features until the Az declined. The optimized subset of observers then were combined using a third perceptron. A subset of 80 features was selected which gave an Az of 0.972. Additionally, at 98.6% sensitivity, the classifier had a specificity of 71.3% and increased the positive predictive value from 60.7% to 84.1 %. Preliminary results suggest that using SRHOs in combination with perceptrons can provide a successful classification scheme for pulmonary nodules. This approach could be incorporated into a larger computer aided detection system for decreasing false positives.

Differences in clinical and laboratory diagnostic characteristics of penicilliosis marneffei in human immunodeficiency virus (HIV)- and non-HIV-infected patients

We compared the clinical and laboratory features of human immunodeficiency virus (HIV)- and non-HIV-infected patients with penicilliosis marneffei. HIV-infected patients had a higher incidence of fungemia. A total of 85.7% of the HIV-negative patients had underlying diseases including hematologic malignancies or had received therapy with corticosteroids or cytotoxic agents. By a Penicillium marneffei-specific mannoprotein Mp1p enzyme-linked immunosorbent assay, serum antigen titers were found to be higher in HIV-positive patients, whereas serum antibody levels were found to be higher in HIV-negative patients.

A neural network approach to breast cancer diagnosis as a constraint satisfaction problem

A constraint satisfaction neural network (CSNN) approach is proposed for breast cancer diagnosis using mammographic and patient history findings. Initially, the diagnostic decision to biopsy was formulated as a constraint satisfaction problem. Then, an associative memory type neural network was applied to solve the problem. The proposed network has a flexible, nonhierarchical architecture that allows it to operate not only as a predictive tool but also as an analysis tool for knowledge discovery of association rules. The CSNN was developed and evaluated using a database of 500 nonpalpable breast lesions with definitive histopathological diagnosis. The CSNN diagnostic performance was evaluated using receiver operating characteristic analysis (ROC). The results of the study showed that the CSNN ROC area index was 0.84+/-0.02. The CSNN predictive performance is competitive with that achieved by experienced radiologists and backpropagation artificial neural networks (BP-ANNs) presented before. Furthermore, the study illustrates how CSNN can be used as a knowledge discovery tool overcoming some of the well-known limitations of BP-ANNs.

Case-based reasoning computer algorithm that uses mammographic findings for breast biopsy decisions

OBJECTIVE

We present case-based reasoning computer software developed from mammographic findings to provide support for the clinical decision to perform biopsy of the breast.

SUBJECTS AND METHODS

The case-based reasoning system is designed to support the decision to perform biopsy in those patients who have suspicious findings on diagnostic mammography. Currently, between 66% and 90% of biopsies are performed on benign lesions. Our system is designed to help decrease the number of benign biopsies without missing malignancies. Clinicians interpret the mammograms using a standard reporting lexicon. The case-based reasoning system compares these findings with a database of cases with known outcomes (from biopsy) and returns the fraction of similar cases that were malignant. This malignancy fraction is an intuitive response that the clinician can then consider when making the decision regarding biopsy.

RESULTS

The system was evaluated using a round-robin sampling scheme and performed with an area under the receiver operating characteristic curve of 0.83, comparable with the performance of a neural network model. If only the cases returning a malignancy fraction of greater than a threshold of 0.10 are sent to biopsy, no malignancies would be missed, and the number of benign biopsies would be decreased by 25%. At a threshold of 0.21, 98%, of the malignancies would be biopsied, and the number of benign biopsies would be decreased by 41%.

CONCLUSION

This preliminary investigation indicates that the case-based reasoning approach to computer-aided diagnosis has the potential to improve the accuracy of breast cancer diagnosis on mammography.

beta-lactamases in Shigella flexneri isolates from Hong Kong and Shanghai and a novel OXA-1-like beta-lactamase, OXA-30

Ninety-one ampicillin-resistant Shigella flexneri strains from Hong Kong and Shanghai were studied for production of beta-lactamases. TEM-1-like and OXA-1-like enzymes were identified in 21 and 79% of the strains, respectively, by isoelectric focusing (IEF). No difference in the pattern of beta-lactamase production was found between strains from Hong Kong and Shanghai. Four ribotypes were detected. Over 88% of OXA-producing strains had the same ribotype. All TEM-1-like strains harbored a plasmid which hybridized positively with the bla(TEM) probe. Total DNA from OXA-1-like strains failed to hybridize or only hybridized weakly with an OXA probe. The OXA resistance was not transferable. OXA-1-like enzymes exhibited substrate and inhibition profiles similar to that of OXA-1 and were shown to have a pI of 7.3 by further IEF using a narrow-range ampholine gel. The gene encoding the OXA-1-like enzyme from one isolate (CH-07) was cloned, sequenced, and found to differ from bla(OXA-1) at codon 131 (AGA-->GGA; Arg to Gly), resulting in the novel designation OXA-30. The predominance of OXA-type enzymes in ampicillin-resistant S. flexneri suggests host preference for specific beta-lactamases.

Segmentation of suspicious clustered microcalcifications in mammograms

We have developed a multistage computer-aided diagnosis (CAD) scheme for the automated segmentation of suspicious microcalcification clusters in digital mammograms. The scheme consisted of three main processing steps. First, the breast region was segmented and its high-frequency content was enhanced using unsharp masking. In the second step, individual microcalcifications were segmented using local histogram analysis on overlapping subimages. For this step, eight histogram features were extracted for each subimage and were used as input to a fuzzy rule-based classifier that identified subimages containing microcalcifications and assigned the appropriate thresholds to segment any microcalcifications within them. The final step clustered the segmented microcalcifications and extracted the following features for each cluster: the number of microcalcifications, the average distance between microcalcifications, and the average number of times pixels in the cluster were segmented in the second step. Fuzzy logic rules incorporating the cluster features were designed to remove nonsuspicious clusters, defined as those with typically benign characteristics. A database of 98 images, with 48 images containing one or more microcalcification clusters, provided training and testing sets to optimize the parameters and evaluate the CAD scheme, respectively. The results showed a true positive rate of 93.2% and an average of 0.73 false positive clusters per image. A comparison of our results with other reported segmentation results on the same database showed comparable sensitivity and at the same time an improved false positive rate. The performance of the CAD scheme is encouraging for its use as an automatic tool for efficient and accurate diagnosis of breast cancer.

A neural network to predict symptomatic lung injury

A nonlinear neural network that simultaneously uses pre-radiotherapy (RT) biological and physical data was developed to predict symptomatic lung injury. The input data were pre-RT pulmonary function, three-dimensional treatment plan doses and demographics. The output was a single value between 0 (asymptomatic) and 1 (symptomatic) to predict the likelihood that a particular patient would become symptomatic. The network was trained on data from 97 patients for 400 iterations with the goal to minimize the mean-squared error. Statistical analysis was performed on the resulting network to determine the model's accuracy. Results from the neural network were compared with those given by traditional linear discriminate analysis and the dose-volume histogram reduction (DVHR) scheme of Kutcher. Receiver-operator characteristic (ROC) analysis was performed on the resulting network which had Az = 0.833 +/- 0.04. (Az is the area under the ROC curve.) Linear discriminate multivariate analysis yielded an Az = 0.813 +/- 0.06. The DVHR method had Az = 0.521 +/- 0.08. The network was also used to rank the significance of the input variables. Future studies will be conducted to improve network accuracy and to include functional imaging data.

An evaluation of susceptibility testing methods for ampicillin-sulbactam using a panel of beta-lactamase-producing bacteria

Bacteria possessing TEM-1-like beta-lactamases are generally regarded as susceptible to ampicillin-sulbactam (SAM), while those harboring OXA-1 enzymes are considered resistant. The current study was undertaken to compare susceptibility testing using various combinations of ampicillin and sulbactam to improve clinical correlation. Members of the Enterobacteriaceae family harboring TEM-1, SHV-1 or OXA-1-like beta-lactamases were tested using the agar dilution method. A substantial proportion of strains harboring OXA-1-like beta-lactamases showed false susceptibility to SAM at the 1:1 ratio or fixed sulbactam concentration of 8 microg/ml. At a fixed sulbactam concentration of 4 microg/ml, the activity of ampicillin-sulbactam appeared to be reduced, with large numbers of TEM-1 producers becoming frankly resistant. Results obtained with the 2:1 ratio exhibited the closest correlation with that obtained by the currently recommended disk diffusion test. However, very major errors were still found between the disk diffusion test and agar dilution test, suggesting the necessity for consideration of a change in criteria for interpretation of disk diffusion test results. In conclusion, SAM susceptibility testing by agar dilution using other than a 2:1 ratio is not recommended and results should be interpreted with caution.

Effect of patient history data on the prediction of breast cancer from mammographic findings with artificial neural networks

RATIONALE AND OBJECTIVES

The authors evaluated the contribution of medical history data to the prediction of breast cancer with artificial neural network (ANN) models based on mammographic findings.

MATERIALS AND METHODS

Three ANNs were developed: The first used 10 Breast Imaging Reporting and Data System (BI-RADS) variables; the second, the BI-RADS variables plus patient age; the third, the BI-RADS variables, patient age, and seven other history variables, for a total of 18 inputs. Performance of the ANNs and the original radiologist's impression were evaluated with five metrics: receiver operating characteristic area index (Az); specificity at given sensitivities of 100%, 98%, and 95%; and positive predictive value.

RESULTS

All three ANNs consistently outperformed the radiologist's impression over all five performance metrics. The patient-age variable was particularly valuable. Adding the age variable to the basic ANN model, which used only the BI-RADS findings, significantly improved Az (P = .028). In fact, replacing all history data with just the age variable resulted in virtually no changes for Az or specificity at 98% sensitivity (P = .324 and P = .410, respectively).

CONCLUSION

Patient age was an important variable for the prediction of breast cancer from mammographic findings with the ANNs. For this data set, all history data could be replaced with age alone.

Correlation of in vitro susceptibility testing results for amoxicillin-clavulanate and ampicillin-sulbactam using a panel of beta-lactamase-producing Enterobacteriaceae

Correlation between in vitro susceptibility results for amoxicillin-clavulanate (AMC) and ampicillin-sulbactam (SAM) was studied using 136 clinical and control strains of Enterobacteriaceae harboring TEM-1, SHV-1 or OXA-1-like beta-lactamases. Determination of minimal inhibitory concentration of antibiotics was performed by agar dilution. The beta-lactamases were initially characterized using isoelectric focusing. Further identification was done by DNA hybridization with or without prior PCR amplification. All strains sensitive to SAM were found to be sensitive also to AMC. In contrast, among those susceptible to AMC, only 50% were sensitive to SAM while 36% gave intermediate results and 14% were resistant. Major differences were found solely among SHV-producers while minor differences occurred mostly among TEM-producers. This phenomenon is probably related to the differential activities of clavulanate and sulbactam against various beta-lactamases. In conclusion, testing of Enterobacteriaceae isolates for susceptibility to AMC and SAM should be performed and reported individually to avoid erroneous designation of susceptibility.

Predictive model for the diagnosis of intraabdominal abscess

RATIONALE AND OBJECTIVES

The authors investigated the use of an artificial neural network (ANN) to aid in the diagnosis of intraabdominal abscess.

MATERIALS AND METHODS

An ANN was constructed based on data from 140 patients who underwent abdominal and pelvic computed tomography (CT) between January and December 1995. Input nodes included data from clinical history, physical examination, laboratory investigation, and radiographic study. The ANN was trained and tested on data from all 140 cases by using a round-robin method and was compared with linear discriminate analysis. A receiver operating characteristic curve was generated to evaluate both predictive models.

RESULTS

CT examinations in 50 cases were positive for abscess. This finding was confirmed by means of laboratory culture of aspirations from CT-guided percutaneous drainage in 38 patients, ultrasound-guided percutaneous drainage in five patients, surgery in five patients, and characteristic appearance on CT scans without aspiration in two patients. CT scans in 90 cases were negative for abscess. The sensitivity and specificity of the ANN in predicting the presence of intraabdominal abscess were 90% and 51%, respectively. Receiver operating characteristic analysis showed no statistically significant difference in performance between the two predictive models.

CONCLUSION

The ANN is a useful tool for determining whether an intraabdominal abscess is present. It can be used to set priorities for CT examinations in order to expedite treatment in patients believed to be more likely to have an abscess.

Treatment of adenovirus-associated haemorrhagic cystitis with ganciclovir

We report a 47-year-old bone marrow transplant recipient with haemorrhagic cystitis caused by adenovirus successfully treated with ganciclovir. This is the first report on the use of ganciclovir for the successful treatment of adenoviral infection. Shell vial culture may be more sensitive than conventional culture in the detection of adenovirus in such patients.

Herpes simplex virus type 2 infection in a 5-year-old boy presenting with recurrent chest wall vesicles and a possible history of herpes encephalitis

A 5-year-old hyperkinetic but otherwise healthy child presented with recurrent irritable vesicles and erosions of the anterior chest wall; they have been apparent since the age of 15 months. Wound swab cultures yielded herpes simplex virus type-2 (HSV-2) and Western blot serology showed past exposure to both HSV-1 and HSV-2. Skin biopsy results further supported a herpes virus infection. Magnetic resonance imaging of the brain showed right temporal lobe atrophy. An evaluation showed no evidence of sexual abuse in the patient but a Western blot assay of the mother's serum for HSV-2 was positive, while the father's was negative. In view of the diagnosis of HSV-2 infection in such a young patient, the possible routes of transmission and the time of acquisition of infection were explored. We believe the most likely route of infection in this child was postnatal, through intimate contact with the mother.

Predicting breast cancer invasion with artificial neural networks on the basis of mammographic features

PURPOSE

To evaluate whether an artificial neural network (ANN) can predict breast cancer invasion on the basis of readily available medical findings (ie, mammographic findings classified according to the American College of Radiology Breast Imaging Reporting and Data System and patient age).

MATERIALS AND METHODS

In 254 adult patients, 266 lesions that had been sampled at biopsy were randomly selected for the study. There were 96 malignant and 170 benign lesions. On the basis of nine mammographic findings and patient age, a three-layer backpropagation network was developed to predict whether the malignant lesions were in situ or invasive.

RESULTS

The ANN predicted invasion among malignant lesions with an area under the receiver operating characteristic curve (Az) of .91 +/- .03. It correctly identified all 28 in situ cancers (specificity, 100%) and 48 of 68 invasive cancers (sensitivity, 71%).

CONCLUSION

The ANN used mammographic features and patient age to accurately classify invasion among breast cancers, information that was previously available only by means of biopsy. This knowledge may assist in surgical planning and may help reduce the cost and morbidity of unnecessary biopsy.

Diffuse nodular lung disease on chest radiographs: a pilot study of characterization by fractal dimension

OBJECTIVE

We present a computer-aided diagnostic technique for identifying nodular interstitial lung disease on chest radiographs. The fractal dimension was used as a numerical measure of image texture on digital chest radiographs to distinguish patients with normal lung from those with a diffuse nodular interstitial abnormality.

MATERIALS AND METHODS

Twenty digitized chest radiographs were classified as normal (n = 10) or as containing diffuse nodular abnormality (n = 10) on the basis of readings assigned according to the classification of the International Labour Organization. Regions of interest (ROIs) measuring 1.28 cm2 were selected from the intercostal spaces of these radiographs. The fractal dimension of these ROIs was estimated by power spectrum analysis. The cases were not subtle.

RESULTS

The fractal dimension provided statistically significant discrimination between normal parenchyma and nodular interstitial lung disease. The area under the receiver operating characteristic curve was 0.90 (+/- 0.02). One operating point provides sensitivity of 88% with a specificity of 80%.

CONCLUSION

The fractal dimension can provide a measure of lung parenchymal texture and shows promise as an element of computer-aided diagnosis, characterization, and follow-up of interstitial lung disease.

Vancomycin and amikacin in cell cultures for virus isolation

Contamination of cell cultures for virus isolation has been increasingly encountered. By reviewing and changing the antimicrobials incorporated in cell culture media, we aim to control this problem. Contaminated cell culture fluids were inoculated for bacterial and fungal isolation, identification and antibacterial susceptibility testing. Based on the above results, vancomycin and amikacin were chosen to replace the penicillin and gentamicin used conventionally. Analysis was carried out on various characteristics of cell culture with respect to antimicrobial change. All contaminating Gram-positive bacteria were susceptible to vancomycin while about 80% of the Gram-negative bacteria were sensitive to amikacin. The new antimicrobial combination was not toxic to cell cultures and both antimicrobials were found to remain stable in media for over six months. The virus isolation rate was maintained after antimicrobial change while the contamination rate was reduced from nearly 10% to 1.5%. We thus conclude that vancomycin and amikacin can well replace the conventional penicillin and gentamicin to be incorporated into maintenance and transport media to control the emerging problem of viral culture contamination.

Structure-function relationship of lipoprotein lipase-mediated enhancement of very low density lipoprotein binding and catabolism by the low density lipoprotein receptor. Functional importance of a properly folded surface loop covering the catalytic center

We examined the structure-function relationship of human lipoprotein lipase (hLPL) in its ability to enhance the binding and catabolism of very low density lipoproteins (VLDL) in COS cells. Untransfected COS cells did not bind to or catabolize normal VLDL. Expression of wild-type hLPL by transient transfection enhanced binding, uptake, and degradation of the VLDL (a property of LPL that we call bridge function). Heparin pretreatment and a monoclonal antibody ID7 that blocks LDL receptor-binding domain of apoE both inhibited binding, and apoE2/E2 VLDL from a Type III hyperlipidemic subject did not bind. However, LDL did not reduce 125I-VLDL binding to the hLPL-expressing cells, whereas rabbit beta-VLDL was an effective competitor. By contrast, LDL reduced uptake and degradation of 125I-VLDL to the same extent as excess unlabeled VLDL or beta-VLDL. These data suggest that binding occurs by direct interaction of VLDL with LPL but the subsequent catabolism of the VLDL is mediated by the LDL receptor. Mutant hLPLs that were catalytically inactive, S132A, S132D, as well as the partially active mutant, S251T, and S172G, gave normal enhancement of VLDL binding and catabolism, whereas the partially active mutant S172D had markedly impaired capacity for the process; thus, there is no correlation between bridge function and lipolytic activity. A naturally occurring genetic variant hLPL, S447-->Ter, has normal bridge function. The catalytic center of LPL is covered by a 21-amino acid loop that must be repositioned before a lipid substrate can gain access to the active site for catalysis. We studied three hLPL loop mutants (LPL-cH, an enzymatically active mutant with the loop replaced by a hepatic lipase loop; LPL-cP, an enzymatically inactive mutant with the loop replaced by a pancreatic lipase loop; and C216S/C239S, an enzymatically inactive mutant with the pair of Cys residues delimiting the loop substituted by Ser residues) and a control double Cys mutant, C418S/C438S. Two of the loop mutants (LPL-cH and LPL-cP) and the control double Cys mutant C418S/C438S gave normal enhancement of VLDL binding and catabolism, whereas the third loop mutant, C216S/C239S, was completely inactive. We conclude that although catalytic activity and the actual primary sequence of the loop of LPL are relatively unimportant (wild-type LPL loop and pancreatic lipase loops have little sequence similarity), the intact folding of the loop, flanked by disulfide bonds, must be maintained for LPL to express its bridge function.

Artificial neural network: improving the quality of breast biopsy recommendations

PURPOSE

To evaluate the performance and inter- and intraobserver variability of an artificial neural network (ANN) for predicting breast biopsy outcome.

MATERIALS AND METHODS

Five radiologists described 60 mammographically detected lesions with the American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) nomenclature. A previously programmed ANN used the BI-RADS descriptors and patient histories to predict biopsy results. ANN predictive performance was compared with the clinical decision to perform biopsy. Inter- and intraobserver variability of radiologists' interpretations and ANN predictions were evaluated with Cohen kappa analysis.

RESULTS

The ANN maintained 100% sensitivity (23 of 23 cancers) while improving the positive predictive value of biopsy results from 38% (23 of 60 lesions) to between 58% (23 of 40 lesions) and 66% (23 of 35 lesions; P < .001). Interobserver variability for interpretation of the lesions was significantly reduced by the ANN (P < .001); there was no statistically significant effect on nearly perfect intraobserver reproducibility.

CONCLUSION

Use of an ANN with radiologists' descriptions of abnormal findings may improve interpretation of mammographic abnormalities.

Computer-aided diagnosis of breast cancer: artificial neural network approach for optimized merging of mammographic features

RATIONALE AND OBJECTIVES

An artificial neural network (ANN) approach was developed for the computer-aided diagnosis of mammography using an optimally minimized number of input features.

METHODS

A backpropagation ANN merged nine input features (age plus eight radiographic findings extracted by radiologists) to predict biopsy outcome as its output. The features were ranked, and more important ones were selected to produce an optimal subset of features.

RESULTS

Given all nine features, the ANN performed with a receiver operator characteristic area under the curve (Az) of .95 +/- .01. Given only the four most important features, the ANN performed with an Az of .96 +/- .01. Although not significantly better than the ANN with all nine features, the ANN with the four optimized features was significantly better than expert radiologists' Az of .90 +/- .02 (p = .01). This four-feature ANN had a 95% sensitivity and an 81% specificity. For cases with calcifications, the radiologists' performance dropped to an Az of .85 +/- .04, whereas a specialized three-feature ANN performed significantly better with an Az of .95 +/- .02 (p = .02).

CONCLUSION

Given only four input features, the ANN predicted biopsy outcome significantly better than did expert radiologists, who also had access to other radiographic and nonradiographic data. The reduced number of features would substantially decrease data entry efforts and potentially improve the ANN's general applicability.

Breast cancer: prediction with artificial neural network based on BI-RADS standardized lexicon

PURPOSE

To determine if an artificial neural network (ANN) to categorize benign and malignant breast lesions can be standardized for use by all radiologists.

MATERIALS AND METHODS

An ANN was constructed based on the standardized lexicon of the Breast Imaging Recording and Data System (BI-RADS) of the American College of Radiology. Eighteen inputs to the network included 10 BI-RADS lesion descriptors and eight input values from the patient's medical history. The network was trained and tested on 206 cases (133 benign, 73 malignant cases). Receiver operating characteristic curves for the network and radiologists were compared.

RESULTS

At a specified output threshold, the ANN would have improved the positive predictive value (PPV) of biopsy from 35% to 61% with a relative sensitivity of 100%. At a fixed sensitivity of 95%, the specificity of the ANN (62%) was significantly greater than the specificity of radiologists (30%) (P < .01).

CONCLUSION

The BI-RADS lexicon provides a standardized language between mammographers and an ANN that can improve the PPV of breast biopsy.

Lipoprotein lipase: role of intramolecular disulfide bonds in enzyme catalysis

Lipoprotein lipase (LPL) catalyzes the hydrolysis of the triacylglycerol component of triacylglycerol-rich lipoproteins. There are 4 cysteine pairs that are completely conserved among LPLs of all species known. We examined the functional importance of each of the cysteine pairs in enzyme catalysis by examining LPLs produced in Cos cells by transfection. Immunoreactive LPL was produced by vectors encoding the wildtype LPL and each of the 4 cysteine-pair mutant LPLs. Enzyme activity was detectable in the wildtype enzyme, but not in 3 of the 4 Cys-->Ser mutant enzymes (C216S/C239S, C264S/C275S, and C278S/C283S). Interestingly, LPL activity was also present in the mutant (C418S/C438S), which affects the C-terminal cysteine pair, with a specific activity approximately 50% higher than that of wildtype. There is evidence that LPL contains two distinct domains consisting of the N-terminal three-quarters of the sequence connected by a flexible region to the C-terminal domain comprising the rest of the molecule. The conservation of catalytic function despite the disruption of the only disulfide bridge in the C-terminal domain of LPL indicates that the two domains can function independently of each other in enzyme catalysis.

Prediction of breast cancer malignancy using an artificial neural network

BACKGROUND

An artificial neural network (ANN) was developed to predict breast cancer from mammographic findings. This network was evaluated in a retrospective study.

METHODS

For a set of patients who were scheduled for biopsy, radiologists interpreted the mammograms and provided data on eight mammographic findings as part of the standard mammographic workup. These findings were encoded as features for an ANN. Results of biopsies were taken as truth in the diagnosis of malignancy. The ANN was trained and evaluated using a jackknife sampling on a set of 260 patient records. Performance of the network was evaluated in terms of sensitivity and specificity over a range of decision thresholds and was expressed as a receiver operating characteristic curve.

RESULTS

The ANN performed more accurately than the radiologists (P < 0.08) with a relative sensitivity of 1.0 and specificity of 0.59.

CONCLUSIONS

An ANN can be trained to predict malignancy from mammographic findings with a high degree of accuracy.

Bayesian restoration of chest radiographs. Scatter compensation with improved signal-to-noise ratio

OBJECTIVES

The authors introduce a Bayesian algorithm for digital chest radiography that increases the signal-to-noise ratio, and thus detectability, for low-contrast objects.

METHOD

The improved images are formed as a maximum a posteriori probability estimation of a scatter-reduced (contrast-enhanced) image with decreased noise. Noise is constrained by including prior knowledge of image smoothness. Variations between neighboring pixels are penalized for small variations (to suppress Poisson noise), but not for larger variations (to avoid affecting anatomical structure). The technique was optimized to reduce residual scatter in digital radiographs of an anatomical chest phantom.

RESULTS

The contrast in the lung was improved by a factor of two, whereas signal-to-noise ratio was improved by a factor of 1.8. Image resolution was unaffected for objects with a contrast greater than 2%.

CONCLUSION

This statistical estimation technique shows promise for improving object detectability in radiographs by simultaneously increasing contrast, while constraining noise.

Scatter compensation in digital chest radiography using the posterior beam stop technique

A new scatter compensation technique for computed radiography based on posterior beam stop (PBS) sampled scatter measurements and the bicubic spline interpolation technique was proposed. Using only a single exposure, both the clinical image and an array of scatter measurements, which were interpolated into a smooth scatter-only image, were simultaneously acquired. The scatter was subtracted from the clinical image to generate the primary-only image. To gauge the accuracy of scatter estimation, both quantitative and interpolation errors were evaluated. The PBS measurements were compared against the standard beam stop method at 16 locations in an anatomical phantom, resulting in quantitative errors of 2.7% relative to the scatter or 6.8% relative to the primary. Also measured were the interpolation error over 64 interpolation sample locations and 64 midpoint sample locations in the anatomical phantom. The combined interpolation error was 1.9% relative to the scatter or 8.0% relative to the primary. At the interpolation sample locations, the errors were identical between the phantom radiograph and digital portable chest radiographs from five patients. By summing the quantitative and interpolation errors in quadrature, the overall error of the PBS SISTER (scatter interpolation-subtraction technique for radiography) method was 3.3% relative to the scatter or 10% relative to the primary, which was adequate for dual-energy imaging purposes (less than 10% error relative to the scatter or 20% relative to the primary). The change of image contrast, noise, and signal-to-noise ratio (SNR) at six locations in the anatomical phantom were quantitatively analyzed. Contrast and noise were equally enhanced in all anatomical regions, resulting in approximately the same SNR before and after compensation.(ABSTRACT TRUNCATED AT 250 WORDS)

Observer evaluation of scatter subtraction for digital portable chest radiographs

RATIONALE AND OBJECTIVES

The authors compared standard digital portable chest radiographs (DPCXR) to scatter-subtracted DPCXR:

METHODS

Thirty DPCXR were obtained using a photostimulable phosphor digital imaging system and a posterior beam stop (PBS) technique that allowed measurement of the scatter component of the DPCXR: The scatter component was subtracted from the clinical image to form a scatter-subtracted image. Six observers recorded preference for the standard image or scatter-subtracted image for identifying five radiographic landmarks and for image quality.

RESULTS

A statistically significant preference was demonstrated for the scatter-subtracted images and for viewing the tracheo-bronchial tree, right paratracheal stripe, vertebral column, and support apparatus position. For unprocessed images, there was a statistically significant preference for viewing the pulmonary vasculature. No statistically significant preference was demonstrated for overall image quality.

CONCLUSIONS

These results suggest that PBS scatter subtraction holds promise for improving visualization of structures in high-scatter regions of chest radiographs.

An artificial neural network for estimating scatter exposures in portable chest radiography

An adaptive linear element (Adaline) was developed to estimate the two-dimensional scatter exposure distribution in digital portable chest radiographs (DPCXR). DPCXRs and quantitative scatter exposure measurements at 64 locations throughout the chest were acquired for ten radiographically normal patients. The Adaline is an artificial neural network which has only a single node and linear thresholding. The Adaline was trained using DPCXR-scatter measurement pairs from five patients. The spatially invariant network would take a portion of the image as its input and estimate the scatter content as output. The trained network was applied to the other five images, and errors were evaluated between estimated and measured scatter values. Performance was compared against a convolution scatter estimation algorithm. The network was evaluated as a function of network size, initial values, and duration of training. Network performance was evaluated qualitatively by the correlation of network weights to physical models, and quantitatively by training and evaluation errors. Using DPCXRs as input, the network learned to describe known scatter exposures accurately (7% error) and estimate scatter in new images (< 8% error) slightly better than convolution methods. Regardless of size and initial shape, all networks adapted into radial exponentials with magnitude of 0.75, perhaps implying an ideal point spread function and average scatter fraction, respectively. To implement scatter compensation, the two-dimensional scatter distribution estimated by the neural network is subtracted from the original DPCXR.

Measurement of scatter fractions in erect posteroanterior and lateral chest radiography

Scatter fractions (SFs) measured in patients undergoing erect posteroanterior (PA) and lateral chest radiography with a 12:1 antiscatter grid are reported. Modifications to the posterior beam-stop (PBS) technique allowed measurement of scatter in these patients, without altering the diagnostic image and without additional radiation exposure. The SF measurements are reported by anatomic location on 42 clinical chest images. Average SF values ranged from 0.27 to 0.90 on lateral radiographs and from 0.27 to 0.68 on PA radiographs. Scatter measurements with the 12:1 grid were found to be greater than estimates from previous PA chest phantom experiments. To the authors' knowledge, they were the first to measure radiation scatter with the PBS technique in patients undergoing PA and lateral chest radiography with the antiscatter grid.

Scatter compensation for digital chest radiography using maximum likelihood expectation maximization

RATIONALE AND OBJECTIVES

An iterative maximum likelihood expectation maximization algorithm (MLEM) has been developed for scatter compensation in chest radiography.

METHODS

The MLEM technique produces a scatter-reduced image which maximizes the probability of observing the measured image. We examined the scatter content and the low-contrast signal-to-noise ratio (SNR) in digital radiographs of anatomical phantoms before and after compensation.

RESULTS

MLEM converged to an accurate (6.4% RMS residual scatter error) estimate within 12 iterations. Both contrast and noise were increased in the processed images as iteration progressed. In the lung, contrast was increased 108% and SNR was improved by 10%. In the retrocardiac region, contrast was increased 180% while SNR decreased by 6%.

CONCLUSIONS

This is the first report of a post-acquisition scatter compensation technique which can increase SNR. These results suggest that statistical estimation techniques can enhance image quality and quantitative accuracy for digital chest radiography.

Thermostable, salt tolerant, wide pH range novel chitobiase from Vibrio parahemolyticus: isolation, characterization, molecular cloning, and expression

A chitobiase gene from Vibrio parahemolyticus was cloned into plasmid pUC18 in Escherichia coli strain DH5 alpha. The plasmid construct, pC120, contained a 6.4 kb Vibrio DNA insert. The recombinant gene expressed chitobiase [EC 3.2.1.30] activity similar to that found in the native Vibrio. The enzyme was purified by ion exchange, hydroxylapatite and gel permeation chromatographies, and exhibited an apparent molecular weight of 80 kDa on SDS-polyacrylamide gel electrophoresis. Chitobiose and 6 more substrates, including beta-N-acetyl galactosamine glycosides, were hydrolyzed by the recombinant chitobiase, indicating its putative classification as an hexosaminidase [EC 3.2.1.52]. The enzyme was resistant to denaturation by 2 M NaCl, thermostable at 45 degrees C and active over a very unusual (for glycosyl hydrolases) pH range, from 4 to 10. The purified cloned chitobiase gave 4 closely focussed bands on an isoelectric focusing gel, at pH 4 to 6.5. The N-terminal 43 amino acid sequence shows no homology with other proteins in commercial databanks or in the literature, and from its N-terminal sequence, appears to be a novel protein, unrelated in sequence to chitobiases from other Vibrios reported and unrelated to hexosaminidases from other organisms.

Production of Biologically Active Recombinant Human Lactoferrin in Aspergillus Oryzae

We report the production of recombinant human lactoferrin in Aspergillus oryzae. Expression of human lactoferrin (hLF), a 78 kD glycoprotein, was achieved by placing the cDNA under the control of the A. oryzae alpha-amylase promoter and the 3' flanking region of the A. niger glucoamylase gene. Using this system, hLF is expressed and secreted into the growth medium at levels up to 25 mg/l. The recombinant lactoferrin is indistinguishable from human milk lactoferrin with respect to its size, immunoreactivity, and iron-binding capacity. The recombinant protein appears to be appropriately N-linked glycosylated and correctly processed at the N-terminus by the A. oryzae secretory apparatus. Lactoferrin is the largest heterologous protein and the first mammalian glycoprotein expressed in the Aspergillus system to date. Hence, this expression system appears suitable for the large-scale production and secretion of biologically active mammalian glycoproteins.

Measurement of scatter fractions in clinical bedside radiography

The authors present measurements of scatter fraction (SF), the ratio of scattered to total imaged photons, from clinical bedside radiographs of 102 patients. These measurements were obtained by using a new posterior beam-stop technique that does not alter the diagnostic image but that simultaneously provides SF measurements at 224 locations in the image. The SF values in the lung were found to be consistent with previous measurements, while the SF values in the mediastinal and retrocardiac areas were larger than previously reported. SFs in diseased lung were significantly larger than SFs in normal lung. The range of SF values was large for all anatomic locations. For applications in which accurate scatter estimation is required, this wide range of values suggests that SFs should be measured in each individual image.

Posterior beam-stop method for scatter fraction measurement in digital radiography

The authors presented a new posterior beam-stop (PBS) technique for measuring the ratio of scattered to total-detected photon flux (scatter fraction) in a radiographic examination while preserving the diagnostic quality of the image. The scatter measurement was made using a standard imaging geometry with both beam stops and an additional x-ray detector placed behind the standard imaging detector. This PBS geometry differs from the standard beam-stop (SBS) technique for scatter measurement. With SBS, a beam-stop shadow appears on the image. To evaluate the PBS technique, scatter fraction measurements were performed on an anatomic phantom using both the PBS and SBS techniques. When compared with the standard technique, PBS provided accurate estimation of scatter fractions. Since the measurement can be performed without degrading a standard clinical radiographic examination, the PBS technique allows simultaneous acquisition of scatter measurements from human patients in combination with a standard radiographic examination.

Quantitative scatter measurement in digital radiography using a photostimulable phosphor imaging system

X-ray scatter fractions measured with two detectors are compared: a photostimulable phosphor system (PSP) and a conventional film-screen technique. For both detection methods, a beam-stop technique was used to estimate the scatter fraction in polystyrene phantoms. These scatter fraction measurements are compared to previously reported film-based measurements. Scatter fractions obtained with the PSP were in good agreement both with measurements using film as well as with most previously reported measurements. For the PSP measurements, repeatability was better than 1%. It was found that the PSP provides a precise x-ray detector for quantitative scatter measurement in digital radiography.

Scatter fractions in AMBER imaging

Images of two phantoms were obtained with use of an advanced multiple-beam equalization radiography system, and scatter fractions were estimated with use of a photostimulable phosphor imaging system. Scatter fractions in the equalized images were lower in the mediastinum-equivalent areas and higher in the lung-equivalent areas, relative to images that were conventionally acquired with use of an antiscatter grid. The differences are attributed to a reduction in incident exposure in the lungs and the presence of cross-scatter between lung and mediastinal regions.