Distinction of cervical cancer biopsies by use of infrared microspectroscopy and probabilistic neural networks.

Fourier-transform infrared spectroscopy has shown alterations of spectral characteristics of cells and tissues as a result of carcinogenesis. The research reported here focuses on the diagnosis of cancer in formalin-fixed biopsied tissue for which immunochemistry is not possible and when PAP-smear results are to be confirmed. The data from two groups of patients (a control group and a group of patients diagnosed with cervical cancer) were analyzed. It was found that the glucose/phosphate ratio decreases (by 23-49%) and the RNA/DNA ratio increases (by 38-150%) in carcinogenic compared with normal tissue. Fourier-transform microspectroscopy was used to examine these tissues. This type of study in larger populations may help to set standards or classes with which to use treated biopsied tissue to predict the possibility of cancer. Probabilistic neural networks and statistical tests as parts of these biopsies predict the possibility of cancer with a high degree of accuracy (> 95%).

Possible common biomarkers from FTIR microspectroscopy of cervical cancer and melanoma.

Detection of malignancy at early stages is crucial in cancer prevention and management. Fourier transform infrared (FTIR) spectroscopy has shown promise as a non-invasive method with diagnostic potential in cancer detection. Studies were conducted with formalin-fixed biopsies of melanoma and cervical cancer by FTIR microspectroscopy (FTIR-MSP) to detect common biomarkers, which occurred in both types of cancer distinguishing them from the respective non-malignant tissues. Both types of cancer are diagnosed on skin surfaces. The spectra were analysed for changes in levels of biomolecules such as RNA, DNA, phosphates and carbohydrate (glycogen). Whereas carbohydrate levels showed a good diagnostic potential for detection of cervical cancer, this was not the case for melanoma. However, variation of the RNA/DNA ratio as measured at I(1121)/I(1020) showed similar trends between non-malignant and malignant tissues in both types of cancer. The ratio was higher for malignant tissues in both types of cancer.

Fourier transform infrared microspectroscopy as a quantitative diagnostic tool for assignment of premalignancy grading in cervical neoplasia.

The early diagnosis and proper identification of cervical squamous intraepithelial lesions plays an important role in a good prognosis for the patient. However, the present practice of screening based on PAP (Papanicolaou) smear and histopathology makes it tedious and prone to human errors. We assess the validity of FTIR microspectroscopy (FTIR-MSP) of biopsies as a method to properly assign the correct stage of premalignancy in patients with symptoms of cervical intraepithelial neoplasia. For the first time we evaluate the biopsies based on the FTIR spectra for different grades of neoplasia in tandem with probabilistic neural networks (PNNs) and histopathology. The results show that the grading of neoplasia based on FTIR-MSP and a PNN differentiates the normal from premalignant with a high level of accuracy. The false positive identification of the normal as cervical intraepithelial neoplasia 1 (CIN1), CIN2, and CIN3 patients is 9.04, 0.01, and 0.01%, respectively. The false negative identification of CIN2 patients as normal and CIN1 patients is 0.01 and 4.4%, respectively. Similarly, the false negative identification of CIN3 patients as normal, CIN1, and CIN2 is 0.14, 6.99, and 9.61%, respectively. The small errors encountered in the grading are comparable to current methods, encouraging advanced studies for the development of mechanized equipment for the diagnosis and grading of premalignant cervical neoplasia.

Optimization of feeding profile for a fed-batch bioreactor by an evolutionary algorithm.

The optimal feeding profile of a fed batch process was designed by means of an evolutionary algorithm. The algorithm chromosomes include the real-valued parameters of a profile function, defined by previous knowledge. Each chromosome is composed of the parameters that define the feeding profile: the feed rates, the singular arc parameters and the switching times between the profile states. The feed profile design was tested on a fed-batch process simulation. The accepted profiles were smooth and similar to those derived analytically in other studies. Two selection functions, roulette wheel and geometric ranking, were compared. In order to overcome the problem of model mismatches, a novel optimization scheme was carried out. During its operation the process was sampled, the model was updated and the optimization procedure was applied. The on-line optimization showed improvement in the objective function for relatively low sample times. Choosing the sampling frequencies depends on the process dynamics and the time required for the measurements and optimization. Further study on experiments of fed-batch process demonstrated the use of complex, non-differentiable model and produced improved process performances using the optimal feeding profile.

Monitoring and control of pullulan production using vision sensor.

The production of the polysaccharide pullulan by the yeast-like fungi, Aureobasidium pullulans, is accompained by cellular morphogenetic changes. High productivity and yield of the process have been found to correlate with high concentration of yeast-like cells in the culture. The morphogenetic changes of A. pullulans cells depend on the culture conditions, e.g., dissolved oxygen, shear rate and medium composition. In order to improve the productivity of the process, a novel control law was formulated. A feeding strategy dependent on the culture cellular composition was designed and aimed to keep the yeast-like cell concentration high. The culture morphogenetic composition during the process was monitored by a recently developed vision sensor. Feeding was actuated when the yeast-like cell concentration decreased below a threshold. The proposed control strategy improved pullulan production by increasing both productivity and yield of the cells by 67% and 80%, correspondingly. The results point to the advantage and the potential of using the monitoring and control system and algorithm to increase productivity and yield in cellular bioprocesses.

Hybrid model building methodology using unsupervised fuzzy clustering and supervised neural networks.

This paper suggests a model building methodology for dealing with new processes. The methodology, called Hybrid Fuzzy Neural Networks (HFNN), combines unsupervised fuzzy clustering and supervised neural networks in order to create simple and flexible models. Fuzzy clustering was used to define relevant domains on the input space. Then, sets of multilayer perceptrons (MLP) were trained (one for each domain) to map input-output relations, creating, in the process, a set of specified sub-models. The estimated output of the model was obtained by fusing the different sub-model outputs weighted by their predicted possibilities. On-line reinforcement learning enabled improvement of the model. The determination of the optimal number of clusters is fundamental to the success of the HFNN approach. The effectiveness of several validity measures was compared to the generalization capability of the model and information criteria. The validity measures were tested with fermentation simulations and real fermentations of a yeast-like fungus, Aureobasidium pullulans. The results outline the criteria limitations. The learning capability of the HFNN was tested with the fermentation data. The results underline the advantages of HFNN over a single neural network.

Unsupervised speaker recognition based on competition between self-organizing maps.

We present a method for clustering the speakers from unlabeled and unsegmented conversation (with known number of speakers), when no a priori knowledge about the identity of the participants is given. Each speaker was modeled by a self-organizing map (SOM). The SOMs were randomly initiated. An iterative algorithm allows the data move from one model to another and adjust the SOMs. The restriction that the data can move only in small groups but not by moving each and every feature vector separately force the SOMs to adjust to speakers (instead of phonemes or other vocal events). This method was applied to high-quality conversations with two to five participants and to two-speaker telephone-quality conversations. The results for two (both high- and telephone-quality) and three speakers were over 80% correct segmentation. The problem becomes even harder when the number of participants is also unknown. Based on the iterative clustering algorithm a validity criterion was also developed to estimate the number of speakers. In 16 out of 17 conversations of high-quality conversations between two and three participants, the estimation of the number of the participants was correct. In telephone-quality the results were poorer.

FT-IR microscopic characterization of normal and malignant human colonic tissues.

Fourier-transform infrared spectroscopy (FT-IR) employs a unique approach to optical diagnosis of tissue pathology based on the characteristic molecular vibrational spectra of the tissue. In this study, we report infrared absorption spectra of formalin fixed, paraffin embedded normal and malignant human colonic tissues from ten different patients. Our method is based on microscopic infrared study (FT-IR-microscopy) of thin tissue specimens in parallel with normal histopathological analysis, which serves as a reference. Our results indicate that the normal colonic tissue has a stronger absorption than the cancerous type over a wide region in all ten cases. The detailed analysis showed that there is a significant decrease in carbohydrate levels, total phosphate and also possibly creatine contents for cancerous tissue in comparison to the controls. Also, RNA/DNA ratio increased in cancerous tissues relative to the normals in all the patients. The results of Linear Discriminant Analysis (LDA) showed that the normal and malignant cells could be identified with about 89% accuracy.

Classification of brain-stem trigeminal evoked potentials in multiple sclerosis, minor head injuries and post-concussion syndrome pathologies by similarity measurements.

In this study measurements obtained from brain-stem trigeminal evoked potentials (BTEP) are applied to the problem of diagnosing Multiple Sclerosis (MS) and Post-concussion syndrome (PCS). We present a simplistic model that depicts the BTEP waveform as the linear combination of a set of filters excited by a short stimulus. The relation between the BTEP latencies and the 1st to 4th harmonic components is shown. The performance of a fuzzy similarity measure based classifier is compared with that of human experts. The efficiency of the proposed classifier in conjunction with delay time and amplitude features is examined. Using this novel approach, a classification rate of 93.55% and 84.1% for MS and PCS pathologies, respectively, was achieved. This performance compares favorably to the classification rates of 84.28% for MS and 70.47% for PCS pathologies achieved by human experts.

Electrical properties of glucose-sensitive hydrogels: swelling and conductivity relationships.

The electrical characteristics of a glucose-sensitive polymeric hydrogel have been studied. The hydrogel matrices were prepared by radical polymerization of solutions containing 2-hydroxyethyl methacrylate, N,N-dimethyl aminoethyl methacrylate, tetraethylene glycol dimethacrylate, ethylene glycol, water, and glucose oxidase. The hydrogels displayed faster and higher swelling rates for lower levels of a crosslinking agent. Electrical conductivity was found to be a sensitive measurement of the state of the swelling. A simple model that relates hydrogel swelling and conductivity has been proposed.

A comparison of neural network and Bayes recognition approaches in the evaluation of the brainstem trigeminal evoked potentials in multiple sclerosis.

This article describes the application of Multi-Layer Perceptron (MLP), Probabilistic Neural Network and Kohonen's Learning Vector Quantization to the problem of diagnosing Multiple Sclerosis. The classification information is obtained from brainstem trigeminal evoked potential. The performance of the neural networks based classifiers is compared with that of the human experts and the Bayes classifier. The ability of the MLP classifier to generalize is far better than that of the Bayes classifier. The efficiency of the neural network based classifiers in conjunction with several types of well-known evoked potential features, such as Fourier transform space, latency and temporal wave, is examined. Although a large clinical data base would be necessary, before this approach can be fully validated, the initial results are promising.

A self-tuning vision system for monitoring biotechnological processes: I. Application to production of pullulan by Aureobasidium pullulans.

A prototype of a self-tuning vision system (STVS) has been developed to monitor cell population in fermentations. The STVS combines classical image processing techniques, neural networks and fuzzy logic technologies. By combining these technologies the STVS is able to analyze sampled images of the culture. The proposed system can be "tailored" with minimum effort by an expert who can "teach" the system to recognize cells by showing examples of different morphologies. After adaptation, the STVS is able to capture images, isolate the different cells, classify them according to the expert's criteria, and provide the profile of the cell's population. The system was applied to the classification and analysis of Aureobasidium pullulans. The importance of understanding the changes of population distribution during the fermentation and its effect in the production of pullulan are emphasized. The STVS can be used for monitoring and control of the cell population in small research fermentors or in large-scale production.

A flat inclined modular photobioreactor for outdoor mass cultivation of photoautotrophs.

A flat inclined modular photobioreactor (FIMP) for mass cultivation of photoautotrophic microorganisms is described. It consists of flat glass reactors connected in cascade facing the sun with the proper tilt angles to assure maximal exposure to direct beam radiation. The optimal cell density in reference to the length of the reactor light path was evaluated, and the effect of the tilt angle on utilization of both direct beam as well as diffuse sunlight was quantitatively assessed. The mixing mode and extent were also optimized in reference to productivity of biomass. The FIMP proved very successful in supporting continuous cultures of the tested species of photoautotrophs, addressing the major criteria involved in design optimization of photobioreactors: Made of fully transparent glass, inclined toward the sun and endowed with a high surface-to-volume ratio, it combines an optimal light path with a vigorous agitation system. The maximal exposure to the culture to solar irradiance as well as the substantial control of temperature facilitate, under these conditions, a particularly high, extremely light-limited optimal cell density. The integrated effects of these growth conditions resulted in record volumetric and areal output rates of Monodus subterraneus, Anabana siamensis, and Spirulina platensis. (c) 1996 John Wiley & Sons, Inc.

Neural-network classifiers for automatic real-world aerial image recognition.

We describe the application of the multilayer perceptron (MLP) network and a version of the adaptive resonance theory version 2-A (ART 2-A) network to the problem of automatic aerial image recognition (AAIR). The classification of aerial images, independent of their positions and orientations, is required for automatic tracking and target recognition. Invariance is achieved by the use of different invariant feature spaces in combination with supervised and unsupervised neural networks. The performance of neural-network-based classifiers in conjunction with several types of invariant AAIR global features, such as the Fourier-transform space, Zernike moments, central moments, and polar transforms, are examined. The advantages of this approach are discussed. The performance of the MLP network is compared with that of a classical correlator. The MLP neural-network correlator outperformed the binary phase-only filter (BPOF) correlator. It was found that the ART 2-A distinguished itself with its speed and its low number of required training vectors. However, only the MLP classifier was able to deal with a combination of shift and rotation geometric distortions.

Human chromosome classification using multilayer perceptron neural network.

A multilayer perceptron (MLP) neural network (NN) has been studied for human chromosome classification. Only 10-20 examples were required for the MLP NN to reach its ultimate performance classifying chromosomes of 5 types. The empirical dependence of the entropic error on the number of examples was found to be highly comparable to the 1/t function. The principal component analysis (PCA) was used, both for network initialization and for feature reduction purposes. The PCA demonstrated the importance of retaining most of the image information whenever small training sets are used. The MLP NN classifier outperformed the Bayes piecewise classifier for all the cases tested. The MLP classifier was found to be almost unsusceptible to the ratio of the number of training vectors to the number of features, whereas the piecewise classifier was highly dependent on this ratio.

Infectious diseases of geriatric inmates.

The number of geriatric inmates is rapidly growing because of more frequent incarceration of older offenders as the number of the elderly in the general population increases nationally. The increase is also due to recent changes in sentencing patterns (e.g., longer sentences and tightened parole) that affect younger, long-term inmates. Geriatric inmates often have chronic medical illnesses that may result in hospitalization for infectious complications. These infectious conditions may be related to factors such as institutionalization (e.g., tuberculosis and influenza), chronic medical illness (e.g., pneumococcal pneumonia), and a history of alcohol or drug use (e.g., hepatitis B virus and retrovirus infection). The epidemiology of these conditions is reviewed. Since infectious complications among geriatric inmates will add stress to a correctional health care system that is already burdened by inmates with AIDS-related illnesses, clinical recognition of these complications and preventive measures are of great importance.

A macromodel for outdoor algal mass production.

A model describing growth of an outdoor algal (Spirulina platensis)culture was developed. The model can simulate biomass production, pH, growth rate, oxygen evolution, and CO(2) fixation rate. It was calibrated and validated against experimental data obtained by a novel automatic data logger/controller instrumentation which can number most vital parameters of the culture including on line estimation of oxygen production rate (OPR). The importance of understanding light distribution through the pond and its effects on the photosynthesis and respiration processes are emphasized. A maximum yield of about 38 g day(-1) m(-2) under optimal conditions is predicted. The present model can also be a useful tool for optimization of algal mass production sites.

On-line optimization of biotechnological processes: I. Application to open algal pond.

A new on-line optimization and control procedure applicable to biotechnological systems for which a precise mathematical model is unavailable has been developed and tested. The proposed approach is based on an online search for optimum operating conditions by an automatic system using a modified simplex algorithm to which several features have been added to permit real time operation. The simplex algorithm is the upper level of a hierarchical software package in which the other levels are cost evaluation, control, data acquisition, and signal processing. The optimization method was tested in a laboratory minipond for the cultivation of Spirulina platensis. The controlled parameters were light intensity, optical density, pH, and temperature. The proposed optimization method can be applied to other biological processes provided that the pertinent variables can be measured and controlled and the cost function can be defined mathematically.

Automatic on-line growth estimation method for outdoor algal biomass production.

An on-line measuring procedure for estimating productivity in outdoor algal cultures was developed and tested experimentally. The procedure is based on a previously described method for on-line measuring net O(2) production rate (OPR). The data obtained by this method was found to correlate well with the conventional procedures for estimation productivity by measuring the changes in biomass concentration in the culture. The new procedure seems to be superior to the latter since it can be carried out in an almost continuous way and can give immediate indication on the productivity. OPR could be used to monitor on-line the photosynthetic and/or respiration activity in small research fermentors or in large-scale open systems outdoors.

An automatic method for on-line estimation of the photosynthetic rate in open algal ponds.

An analytical model for dissolved oxygen concentration in an algal minipond was used to develop a new method for estimating, on-line, the net O(2) production rate (OPR) of the biological process. The method was tested experimentally and was found to provide crucial information on the vitality of the biological process and to provide an early warning of a possible forthcoming collapse of the ecosystern. It is suggested that the newly developed model and measurement method could provide investigators with useful tools for optimization of algal cultivation in the laboratory and plant.