Assessing speech intelligibility of pathological speech in sentences and word lists: The contribution of phoneme-level measures

INTRODUCTION

Speech intelligibility is an important indicator of the degree of speech impairment in pathological speech. Articulation, as a key feature of dysarthria, has been found to be a stronger contributor to intelligibility of dysarthric speech compared to voice quality, nasality, and prosody. In fact, therapy addressing articulation is often used by speech-language pathologists. Since phoneme-level measures are more directly related to articulation, they may contribute to better evaluating articulation imprecision in speakers with dysarthria and to monitoring the effectiveness of therapy.

METHOD

We collected two types of phoneme-level measures: a) Accuracy of Phonemes, the percentage of correctly transcribed phonemes, and b) Phonetic Distance, from orthographic transcriptions obtained from expert raters in two types of speech materials (i.e., meaningful sentences and word lists). We first examined the measures' interrater reliability using Generalizability Theory. Then we studied the validity of the measures by correlating them to three criterion variables. Following this, we explored their ability in distinguishing speakers in two classification tasks according to speakers' types (i.e., healthy vs dysarthric) and their severity levels of dysarthria, respectively.

RESULTS

The results showed that both types of phoneme-level measures are highly reliable and valid in two different speech materials. They also showed acceptable results for both classification tasks in different speech materials, with word lists performing better than meaningful sentences. The differences between the two speech materials may be largely caused by differences in word structures and contextual cues in the materials.

CONCLUSION

The results indicate that both types of phoneme-level measures show largely similar reliability and validity in both speech materials. These measures perform better in word lists than in meaningful sentences, suggesting an advantage for using word lists in clinical practice and research. On the other hand, meaningful sentences can be used for classifying healthy and dysarthric speakers. Our results suggest that using different speech materials gives a better overview of the speakers' intelligibility at the segmental level and the implications of their articulation impairments.

Assessing speech intelligibility of pathological speech: test types, ratings and transcription measures

Speech intelligibility is an essential though complex construct in speech pathology. In this paper, we investigated the interrater reliability and validity of two types of intelligibility measures: a rating-based measure, through Visual Analogue Scales (VAS), and a transcription-based measure called Accuracy of Words (AcW), through two forms of orthographic transcriptions, one containing only existing words (EWTrans) and one allowing all sorts of words, including both existing words and pseudowords (AWTrans). Both VAS and AcW scores were collected from five expert raters. We selected speakers with various severity levels of dysarthria (SevL) and employed two types of speech materials, i.e. meaningful sentences and word lists. To measure reliability, we applied Generalizability Theory, which is relatively unknown in the field of pathological speech and language research but enables more comprehensive analyses than traditional methods, e.g., the intraclass correlation coefficient. The results convincingly indicate that five expert raters were sufficient to provide reliable rating-based (VAS) and transcription-based (AcW) measures, and that reliability increased as the number of raters or utterances increased. Generalizability Theory has proved effective in systematically dealing with reliability issues in our experimental design. We also investigated construct and concurrent validity. Construct validity was addressed by exploring the correlations between VAS and AcW within and across speech materials. Concurrent validity was addressed by exploring the correlations between our measures, i.e. VAS and AcW, and two external measures, i.e. phoneme intelligibility and SevL. The correlations corroborate the validity of VAS and AcW to assess speech intelligibility, both in sentences and word lists.

A serious game for speech training in dysarthric speakers with Parkinson's disease: Exploring therapeutic efficacy and patient satisfaction

BACKGROUND

The increasing need for speech therapy due to our ageing population raises the demand on therapeutical resources. To meet this demand, innovative delivery of speech training is required. eHealth applications may provide a solution, as intensified and prolonged training is only possible and affordable in patients' home environment.

AIMS

This study explores the effects on speech intelligibility of game-based speech training that provides automatic feedback on loudness, pitch and pronunciation. Additionally, we investigate how satisfied patients are with the game-based speech training and how they experience the automatic feedback. Furthermore, patients' preferences for game-based speech training compared with face-to-face training are explored.

METHODS AND PROCEDURES

Eight adult dysarthric speakers with Parkinson's disease (PD) completed a 4-week game-based speech training in their home environment. For each speaker, 24 speech utterances were audio recorded 4 weeks before (T1), immediately before (T2) and immediately after (T3) the training. All speech samples were rated on speech intelligibility by 10 untrained listeners, by comparing them with the corresponding utterances realized by a healthy speaker. Changes over time were analysed using a linear mixed-effects analysis. Patient satisfaction with the game and the automatic feedback was assessed using a questionnaire. The preferences of patients were collected using a paired comparisons procedure in which the patients were asked whether they would prefer game-based or face-to-face speech training in four hypothetical scenarios with different hypothesized levels of speech improvement.

OUTCOMES AND RESULTS

While there was no significant difference in speech intelligibility ratings between T1 and T2, we did find one between T2 and T3. At T3, speech intelligibility was rated higher than at T2, indicating positive effects of the game-based speech training. Patients generally seemed satisfied with the game as average ratings were above 7 on a 10-point rating scale. Generally, patients agreed with the automatic feedback and could use it to positively change the way they spoke. Patients prefer the training that provides the highest hypothetical improvement, and thus do not prefer face-to-face above game-based therapy.

CONCLUSIONS AND IMPLICATIONS

The results of this study suggest that dysarthric speakers due to PD see game-based speech therapy as a valid alternative for face-to-face therapy and that it leads to an average improvement in speech intelligibility. For an optimal effect and user satisfaction it should preferably not be used in isolation but in combination with face-to-face training. In this manner, the strengths of both therapeutic deliveries can be harnessed.

WHAT THIS PAPER ADDS

What is already known on this subject Dysarthric speech in patients with PD is known to benefit from intensified and long-term speech therapy. The increasing need for speech therapy due to our ageing population raises the demand on therapeutical resources making highly frequent and long-term therapy difficult. eHealth provides the opportunity to intensify and prolong speech training in patients' home environment. A drill-and-practice method was employed and investigated in a web-based speech application, indicating positive effects on speech intelligibility. However, participants indicated a lack of variation in exercises making the training less enjoyable. Other research showed that serious games can increase enjoyment during training. What this paper adds to existing knowledge The results of this study show that it is possible to develop a serious game that can be successfully used for speech training by patients with dysarthria due to PD. Comparing speech intelligibility ratings before and after training, we observed significant improvements in speech intelligibility ratings. Patients generally agree with the automatic feedback and can use it to positively change the way they speak. Average ratings were above 7 on a 10-point scale, indicating that patients are satisfied with the game. Patients prefer the type of training (game based or face to face) that provides the highest hypothetical improvement. What are the potential or actual clinical implications of this work? The results of this study suggest that game-based speech training can improve speech intelligibility in patients. This indicates that it can be considered a suitable approach in the treatment of patients with dysarthria due to PD. Patients do not prefer game-based training above face-to-face training in all scenarios. For that reason, for every patient one should carefully consider how to optimally combine game-based and face-to-face training.

Individual word activation and word frequency effects during the processing of opaque idiomatic expressions

Idiom processing studies have paid considerable attention to the relationship between idiomatic expressions as a whole and their constituent words. Although most research focused on the semantic properties of the constituent words, their orthographic form could also play a role in processing. To test this, we assessed both form and meaning activation of individual words during the processing of opaque idioms. In two primed word naming experiments, Dutch native speakers silently read sentences word by word and then named the last word of the sentence. This target word was embedded in either an idiomatic or a literal context and was expected and correct in this context (COR), semantically related (REL) to the expected word, or unrelated (UNREL) to the expected word. The correct target word in the idiomatic context was always part of an opaque idiom. Faster naming latencies for the idiom-final noun than for the unrelated target in the idiomatic context indicated that the idiom was activated as a whole during processing. In addition, semantic facilitation was observed in the literal context (COR < REL < UNREL), but not in the idiomatic context (COR < REL = UNREL). This is evidence that the idiom-final noun was not activated at the meaning level of representation. However, an inhibitory effect of orthographic word frequency of the idiom-final noun indicated that the idiom-final noun was activated at the form level. These results provide evidence in favour of a hybrid model of idiom processing in which the individual words and the idiom as a whole interact on form and meaning levels of representation.

Deep learning architectures for estimating breathing signal and respiratory parameters from speech recordings

Respiration is an essential and primary mechanism for speech production. We first inhale and then produce speech while exhaling. When we run out of breath, we stop speaking and inhale. Though this process is involuntary, speech production involves a systematic outflow of air during exhalation characterized by linguistic content and prosodic factors of the utterance. Thus speech and respiration are closely related, and modeling this relationship makes sensing respiratory dynamics directly from the speech plausible, however is not well explored. In this article, we conduct a comprehensive study to explore techniques for sensing breathing signal and breathing parameters from speech using deep learning architectures and address the challenges involved in establishing the practical purpose of this technology. Estimating the breathing pattern from the speech would give us information about the respiratory parameters, thus enabling us to understand the respiratory health using one's speech.

Normative Data of Dutch Idiomatic Expressions: Subjective Judgments You Can Bank on

The processing of idiomatic expressions is a topical issue in empirical research. Various factors have been found to influence idiom processing, such as idiom familiarity and idiom transparency. Information on these variables is usually obtained through norming studies. Studies investigating the effect of various properties on idiom processing have led to ambiguous results. This may be due to the variability of operationalizations of the idiom properties across norming studies, which in turn may affect the reliability of the subjective judgements. However, not all studies that collected normative data on idiomatic expressions investigated their reliability, and studies that did address the reliability of subjective ratings used various measures and produced mixed results. In this study, we investigated the reliability of subjective judgements, the relation between subjective and objective idiom frequency, and the impact of these dimensions on the participants' idiom knowledge by collecting normative data of five subjective idiom properties (Frequency of Exposure, Meaning Familiarity, Frequency of Usage, Transparency, and Imageability) from 390 native speakers and objective corpus frequency for 374 Dutch idiomatic expressions. For reliability, we compared measures calculated in previous studies, with the D-coefficient, a metric taken from Generalizability Theory. High reliability was found for all subjective dimensions. One reliability metric, Krippendorff's alpha, generally produced lower values, while similar values were obtained for three other measures (Cronbach's alpha, Intraclass Correlation Coefficient, and the D-coefficient). Advantages of the D-coefficient are that it can be applied to unbalanced research designs, and to estimate the minimum number of raters required to obtain reliable ratings. Slightly higher coefficients were observed for so-called experience-based dimensions (Frequency of Exposure, Meaning Familiarity, and Frequency of Usage) than for content-based dimensions (Transparency and Imageability). In addition, fewer raters were required to obtain reliable ratings for the experience-based dimensions. Subjective and objective frequency appeared to be poorly correlated, while all subjective idiom properties and objective frequency turned out to affect idiom knowledge. Meaning Familiarity, Subjective and Objective Frequency of Exposure, Frequency of Usage, and Transparency positively contributed to idiom knowledge, while a negative effect was found for Imageability. We discuss these relationships in more detail, and give methodological recommendations with respect to the procedures and the measure to calculate reliability.

[Cytology of cerebrospinal fluid : Standards, importance and modern methods]

Cytology is an integral part of cerebrospinal fluid (CSF) analysis. It is relevant for the diagnostics and differential diagnosis of inflammatory, hemorrhagic and neoplastic central nervous system (CNS) processes. This article summarizes the recommended procedures and typical clinical patterns. In addition, modern immunocytochemical and flow cytometry methods for CSF cytology are presented. In particular, the diagnostic contribution and clinical relevance in several CNS conditions are discussed.

Roscovitine has anti-proliferative and pro-apoptotic effects on glioblastoma cell lines: A pilot study

Purine analogue roscovitine, a cyclin-dependent kinase (CDK) inhibitor, has shown strong anti-proliferative and pro-apoptotic effects in solid and hematologic cancers such as non small-cell lung cancer and lymphomas. It targets CDK2, 7 and 9 preferentially, which are also overexpressed in glioblastoma. Τherefore, the biological effects of roscovitine in glioblastoma cell lines were investigated. Glioblastoma A172 and G28 cell lines were incubated with serial concentrations of roscovitine for 24-120 h. Proliferation was measured using the xCELLigence Real-Time Cell Analyzer, an impedance‑based cell viability system. Cell cycle distribution was assessed by flow cytometry and gene expression was quantified by quantitative RT-PCR and western blot analysis. Roscovitine exhibited a clear dose-dependent anti‑proliferative and pro‑apoptotic effect in the A172 cell line, while G28 cells showed a anti-proliferative effect only at 100 µM. The results of the flow cytometric (FACS) analysis revealed a dose-dependent increase of the G2/M and sub-G1 fractions in A172 cells, while G28 cells responded with an elevated sub-G1 fraction only at the highest concentration. Roscovitine led to a dose‑dependent decrease of transcripts of p53, CDK 7 and cyclins A and E and an increase of >4-fold of p21 in A172 cells. In G28 cells, a dose‑dependent induction of CDK2, p21 and cyclin D was observed between 10 and 50 µM roscovitine after 72 h, however, at the highest concentration of 100 µM, all investigated genes were downregulated. Roscovitine exerted clear dose-dependent anti-proliferative and pro-apoptotic effects in A172 cells and less distinct effects on G28 cells. In A172 cells, roscovitine led to G2/M arrest and induced apoptosis, an effect accompanied by induced p21 and a reduced expression of CDK2, 7 and 9 and cyclins A and E. These effects requre further studies on a larger scale to confirm whether roscovitine can be used as a therapeutic agent against glioblastoma.

Automatic pronunciation error detection in non-native speech: the case of vowel errors in Dutch

This research is aimed at analyzing and improving automatic pronunciation error detection in a second language. Dutch vowels spoken by adult non-native learners of Dutch are used as a test case. A first study on Dutch pronunciation by L2 learners with different L1s revealed that vowel pronunciation errors are relatively frequent and often concern subtle acoustic differences between the realization and the target sound. In a second study automatic pronunciation error detection experiments were conducted to compare existing measures to a metric that takes account of the error patterns observed to capture relevant acoustic differences. The results of the two studies do indeed show that error patterns bear information that can be usefully employed in weighted automatic measures of pronunciation quality. In addition, it appears that combining such a weighted metric with existing measures improves the equal error rate by 6.1 percentage points from 0.297, for the Goodness of Pronunciation (GOP) algorithm, to 0.236.

Glioma infiltration of the corpus callosum: early signs detected by DTI

The most frequent primary brain tumors, anaplastic astrocytomas (AA) and glioblastomas (GBM): tend to invasion of the surrounding brain. Histopathological studies found malignant cells in macroscopically unsuspicious brain parenchyma remote from the primary tumor, even affecting the contralateral hemisphere. In early stages, diffuse interneural infiltration with changes of the apparent diffusion coefficient (ADC) and fractional anisotropy (FA) is suspected. The purpose of this study was to investigate the value of DTI as a possible instrument of depicting evidence of tumor invasion into the corpus callosum (CC). Preoperatively, 31 patients with high-grade brain tumors (8 AA and 23 GBM) were examined by MRI at 3 T, applying a high-resolution diffusion tensor imaging (DTI) sequence. ADC- and FA-values were analyzed in the tumor-associated area of the CC as identified by fiber tracking, and were compared to matched healthy controls. In (MR-)morphologically normal appearing CC the ADC values were elevated in the tumor patients (n = 22; 0.978 × 10⁻³ mm²/s) compared to matched controls (0.917 × 10⁻³ mm²/s, p < 0.05), and the corresponding relative FA was reduced (rFA: 88 %, p < 0.01). The effect was pronounced in case of affection of the CC visible on MRI (n = 9; 0.978 × 10⁻³ mm²/s, p < 0.05; rFA: 72 %, p < 0.01). Changes in diffusivity and anisotropy in the CC can be interpreted as an indicator of tumor spread into the contralateral hemisphere not visible on conventional MRI.

[Primary brain tumors and brain metastases. Symptomatic epilepsy and driving ability - systematic review and expert opinion]

PURPOSE

Primary brain tumors and metastases are common causes of symptomatic epilepsy. Seizures, neurological and neuropsychological deficits can interfere with driving ability. The present paper aims to systematically review the incidence of epileptic seizures in brain tumor patients and to discuss driving ability in the context of the current German guidelines and expert opinions.

METHODS

To evaluate the incidence of epileptic seizures which occur at the beginning and in the course of the disease, we performed a systematic literature research in PubMed from 1960 to 2007. Additionally on the basis of this data we performed a survey collecting expert opinions regarding the driving ability of brain tumor patients from members of the German working groups "Arbeitsgemeinschaft für prächirurgische Epilepsiediagnostik und operative Epilepsietherapie" (Working Group for Presurgical Epilepsy Diagnostics and Operative Epileptic Therapy) and "Neuroonkologische Arbeitsgemeinschaft" (Neuro-oncological Working Group).

RESULTS

The incidence of epileptic seizures depends on the entity, dignity and localization of the tumor. The driving ability of brain tumor patients is not explicitly regulated in Germany. Of the interviewed experts 72% judged the guidelines to be precise enough and 44% did not want to deprive the patients of their driving ability without a first seizure, independent of the individual risk.

DISCUSSION

The available studies are methodologically insufficient and show that a further evaluation is necessary to assess the driving ability. Possible restrictions of the driving ability in patients with a high risk of seizures in the course of the disease have to take into account the balance between individual rights and the interests of the general public.

High expression levels of erythropoietin and its receptor are not correlated with shorter survival in human glioblastoma

Erythropoietin (EPO) is used to treat anemia in neoplastic disease. EPO also exerts neuroprotective effects on neuronal cells, making a prophylactic use against the neurocognitive effects of radiochemotherapy probable. However, EPO/EPO-receptor (EPOR) signalling has been also detected in glioblastoma cells. Data collected in vitro and in vivo show conflicting results on the effect of EPO on malignant gliomas. The association between EPO and EPOR expression and the prognosis of human glioblastomas was analyzed. Probes of human glioblastomas with complete documentation of clinical course and treatment were assessed by immunohistochemistry for the expression of EPO and EPOR (n=80). Using univariate and multivariate survival analysis, the association with age, gender, radiation, chemotherapy and extent of resection was determined. High levels of EPOR were correlated with a median survival advantage of 7 months (p<0.01). By univariate, but not multivariate, analysis, high levels of EPO and EPOR were associated with a significant prolongation of 7 months median survival when compared to low levels of both molecules. In patients treated with radiochemotherapy adjuvant to surgery, the median survival was 6.5 months longer in patients with high levels of EPOR (p<0.04). According to previous studies, longer patient survival is associated with EPOR expression. Therefore, EPO appears to be safe for the treatment of anemia in glioblastoma patients. However, a prophylactic use, i.e., for neuroprotection, is not recommended in light of the functional studies described in the literature.

Carcinoembryonic antigen in the CSF of cancer patients--the value of intrathecal synthesis and correlation with IgA-diffusion dynamics

OBJECTIVE

The diagnostic impact of carcinoembryonic antigen (CEA) was evaluated in serum and CSF of cancer and control patients.

METHODS

97 analyses of CEA in CSF and serum from 83 cancer patients were compared with 41 cases without malignancy. CEA diffusion dynamics were evaluated with IgA CSF/serum quotients (Q IgA). Intrathecal synthesis of CEA was analysed both by calculating an index Q CEA/Q IgA and within the IgA-diagram.

RESULTS

In 73 samples without synthesis of IgA or CEA, both quotients correlated well with a mean Q CEA/Q IgA of 1.1 (95% CI 0.97-1.2). The Q CEA/Q IgA was significantly higher in metastasizing adenocarcinomas than in controls or other malignancies. In leptomeningeal disease from adenocarcinoma, Q CEA/Q IgA was significantly higher than in controls, while patients with CNS and/or bone metastases had intermediate values. The sensitivity to detect leptomeningeal disease was 91% and 69% for brain metastases. Q CEA/Q IgA and CEA synthesis assessed with the IgA diagram were equally sensitive.

CONCLUSIONS

Evaluation of CEA in the IgA diagram is feasible and of clinical value. The consideration of intrathecal CEA synthesis correlates better with the clinical status than absolute CSF-CEA or the correlation with albumin.

Neoplastic meningitis from breast cancer: feasibility and activity of long-term intrathecal liposomal Ara-C combined with dose-dense temozolomide

BACKGROUND

Patients with neoplastic meningitis (NM) from breast cancer have a median survival of 4-8 months with specific treatment. Here, good tolerance and long-term stabilization with combined intrathecal liposomal cytarabine (Ara-C), which is probably the most promising drug for intrathecal chemotherapy to date, near-continuous temozolomide and radiotherapy is reported in two patients with leptomeningeal and solid central nervous system (CNS) metastases from breast cancer.

CASE REPORTS

A 42- and a 43-year-old female presented with NM and disseminated CNS metastases from human epidermal growth factor receptor type 2 (Her2)-positive breast cancer. After irradiation of the symptomatic sites, intrathecal liposomal Ara-C every 2-4 weeks was combined with temozolomide 100 mg/m(2) day 1-5/7. Cerebrospinal fluid (CSF) cytology and neurological symptoms improved in both patients and stabilized for several months. The patients survived 10 and 17 months after diagnosis of NM, without signs of neurological toxicity.

CONCLUSION

Intensive treatment is complicated by extensive pre-treatment and the lack of active CNS-penetrating systemic drugs. The long-term results with up to 17 intrathecal injections of liposomal Ara-C show that this treatment regimen is feasible and well-tolerated. The stabilization of both patients indicates activity of this combined intrathecal and systemic regimen that is based on long-term exposure of the tumour cells to both Ara-C and temozolomide. The results need to be confirmed prospectively.

Erythropoietin receptor is expressed in meningiomas and lower levels are associated with tumour recurrence

AIMS

The Epo-EpoR pathway plays a role in tumour growth, metastasis and treatment resistance and is a potential target in oncological treatment. As the EpoR status in human meningiomas is unknown, our aim was to characterize EpoR expression in these tumours.

METHODS

We examined 131 meningioma samples of all WHO grades from 116 patients by immunohistochemistry for EpoR. Among these, 25 meningiomas showed brain invasion and 29 patients had a further tumour recurrence. A group of 20 patients without tumour recurrence served as controls. In 12 cases we were able to compare both the primary and the following recurrent tumours. The presence of EpoR in meningiomas was confirmed by RT-PCR and Western blot.

RESULTS

EpoR was expressed in all meningiomas. Statistical analysis revealed that the mean expression levels of EpoR were significantly lower in primary tumours with known recurrence compared with a recurrence-free control group. Additional matched pair analysis in individual cases showed no significant differences between primary and recurrent tumours. No significant correlation between EpoR expression and WHO grade, age, sex or brain invasion was detected. Using specific primer pairs for RT-PCR, we were able to detect all three known isoforms of EpoR: the full-length isoform EpoR-F, the truncated isoform EpoR-T and the soluble isoform EpoR-S.

CONCLUSIONS

Our results demonstrate the expression of EpoR in meningiomas. Lower EpoR mean levels might be a useful marker for a higher recurrence risk, but further studies are needed to clarify the influence of EpoR on recurrences and the role of the different isoforms.

Bag-1 and Bcl-2 gene transfer in malignant glioma: modulation of cell cycle regulation and apoptosis

Bag-1 is a heat shock 70 kDa (Hsp70)-binding protein that can collaborate with Bcl-2 in suppressing apoptosis under some conditions. Here, we report that 11 of 12 human glioma cell lines express Bag-1 protein in vitro. Moreover, 15 of 19 human glioblastomas expressed Bag-1 as assessed by immunohistochemistry in primary tumor specimens. To examine the biological effects of Bag-1 in glioma cells, we expressed Bag-1 or Bcl-2 transgenes in 2 human malignant glioma cell lines, LN-18 and LN-229. Bag-1 significantly slowed glioma cell growth and reduced clonogenicity of both cell lines in vitro. Coexpressed Bcl-2 abrogated these effects of Bag-1. Intracranial LN-229 glioma xenografts implanted into nude mice revealed a substantial growth advantage afforded by Bcl-2. Bag-1 had no such effect, either in the absence or presence of Bcl-2. Upon serum starvation in vitro, Bcl-2 prevented cell death whereas Bag-1 did not. Both Bcl-2 and Bag-1 slowed proliferation of serum-starved cells when expressed alone. Importantly, coexpression of Bcl-2 and Bag-1 provided a distinct growth advantage under conditions of serum starvation that is probably the result of (i) the death-preventing activity of Bcl-2 and (ii) the property of Bag-1 to overcome a Bcl-2-mediated enhancement of exit from the cell cycle. In contrast to these Bcl-2/Bag-1 interactions observed under serum starvation conditions, Bag-1 did not further enhance the strong protection from staurosporine-, CD95 (Fas/Apo1) ligand-, Apo2 ligand (TRAIL)- or chemotherapeutic drug-induced apoptosis afforded by Bcl-2. Taken together, these results indicate a role for Bag-1/Bcl-2 interactions in providing a survival advantage to cancer cells in a deprived microenvironment that may be characteristic of ischemic/hypoxic tumors such as human glioblastoma multiforme, and suggest that Bcl-2/Bag-1 interactions also modulate cell proliferation.

Quantitative assessment of second language learners' fluency: comparisons between read and spontaneous speech

This paper describes two experiments aimed at exploring the relationship between objective properties of speech and perceived fluency in read and spontaneous speech. The aim is to determine whether such quantitative measures can be used to develop objective fluency tests. Fragments of read speech (Experiment 1) of 60 non-native speakers of Dutch and of spontaneous speech (Experiment 2) of another group of 57 non-native speakers of Dutch were scored for fluency by human raters and were analyzed by means of a continuous speech recognizer to calculate a number of objective measures of speech quality known to be related to perceived fluency. The results show that the objective measures investigated in this study can be employed to predict fluency ratings, but the predictive power of such measures is stronger for read speech than for spontaneous speech. Moreover, the adequacy of the variables to be employed appears to be dependent on the specific type of speech material investigated and the specific task performed by the speaker.

Macrophage/microglial cell subpopulations in glioblastoma multiforme relapses are differentially altered by radiochemotherapy

Following surgical removal of glioblastoma multiforme (GBM), radiochemotherapy impedes neoplastic outgrowth and relapse formation. Macrophages/microglial cells are believed to be potent mediators of the host defense system in GBM. However, little is known about their alteration by postsurgical therapies. We have now analyzed expression of LCA (leucocyte common antigen), CD68 (phagocytic cells), HLA-DR, -DP, -DQ (MHC class II), MRP-8 (myeloid-related protein, S100A8), MRP-14 (S100A9), LCF (lymphocyte chemoattractant factor, IL-16) and NOS II (inducible nitric oxide synthase) in macrophages/microglial cells in 39 GBM relapses and their matched primary tumors. Following surgery of the primary tumors, 15 patients received irradiation and chemotherapy, 17 irradiation and 7 no treatment. In irradiated relapses, we observed significantly more macrophages/microglial cells expressing MRP-14 compared to untreated GBM relapses. Furthermore, we observed a significant increase of CD68 expressing macrophages/microglial cells in patients without postsurgical treatment, but not in those with radiochemotherapy. In conclusion, our findings suggest that radiochemotherapy alters the number of MRP-14 expressing cells. The lacking increase of CD68 expressing cells in patients with radiochemotherapy suggests depletion of this cell type by postsurgical therapy.

Obtaining phonetic transcriptions: a comparison between expert listeners and a continuous speech recognizer

In this article, we address the issue of using a continuous speech recognition tool to obtain phonetic or phonological representations of speech. Two experiments were carried out in which the performance of a continuous speech recognizer (CSR) was compared to the performance of expert listeners in a task of judging whether a number of prespecified phones had been realized in an utterance. In the first experiment, nine expert listeners and the CSR carried out exactly the same task: deciding whether a segment was present or not in 467 cases. In the second experiment, we expanded on the first experiment by focusing on two phonological processes: schwa-deletion and schwa-insertion. The results of these experiments show that significant differences in performance were found between the CSR and the listeners, but also between individual listeners. Although some of these differences appeared to be statistically significant, their magnitude is such that they may very well be acceptable depending on what the transcriptions are needed for. In other words, although the CSR is not infallible, it makes it possible to explore large datasets, which might outweigh the errors introduced by the mistakes the CSR makes. For these reasons, we can conclude that the CSR can be used instead of a listener to carry out this type of task: deciding whether a phone is present or not.

Glutathione S-transferase and drug sensitivity in malignant glioma

Malignant gliomas are largely resistant to current approaches of adjuvant chemotherapy. Gluthatione S-transferases (GST) have been attributed a role in the resistance of cancer cells, e.g., to nitrosoureas. Here, we assessed the expression levels of GST-pi and GST-mu RNA and protein as well as total GST activity in a panel of 12 human glioma cell lines and correlated these data with p53 status, BCL-2 family protein expression and drug sensitivity in these cells. Neither GST protein levels nor GST activity correlated with genetic or functional p53 status or with the expression of various BCL-2 family proteins. No evidence for GST-mediated protection from chemotherapeutic drugs became apparent. In contrast, high levels of GST-pi protein, probably the major source of GST activity in glioma cells, and of total GST activity correlated with enhanced sensitivity to vincristine-induced clonogenic cell death. Expression of GST-pi in human glioblastomas in vivo was confirmed by immunohistochemistry. Neither total, nor cytoplasmic or nuclear, GST-pi immunoreactivity correlated with the response to adjuvant radiotherapy or radiochemotherapy. A comparative analysis of primary and recurrent tumors showed that GST expression was not enhanced by radiochemotherapy in vivo. We conclude that GST does not account for the differential chemosensitivity of glioma cell lines in vitro and does not accumulate in glioma subpopulations that form recurrent tumors after radiochemotherapy in vivo.

[Walker-Warburg syndrome in adulthood?]

BACKGROUND

Walker-Warburg-Syndrome (WWS) comprises malformations of the brain and the eyes with muscle dystrophy. Its course is usually lethal within the first 60 weeks of life. As far as we know there are no reports in the literature dealing with WWS in adults.

CASE REPORT

A 70-year old female died because of cardio-pulmonary failure while she had an infection of the respiratory tract. The patient had been mentally deficient and had suffered from recurrent single convulsions. 17 years before death funduscopy had revealed an optic atrophy. Diabetes mellitus was known for ca. 15 years. The neuropathologists found a lissencephaly with a typical histology and a cerebellar malformation. Histologically, the eyes showed diabetic changes like vacuolization of the iris pigment epithelium and a thickening of basement membranes. There were marked deposits in the Pars plicata of the ciliary body and drusen formation in Bruch's membrane. The nerve fiber layer of the retina seemed to be thinned, and there was a considerable rarefication of optic ganglion cells.

CONCLUSIONS

Taking the clinical and especially the morphological data into account, a Walker-Warburg-Syndrome was assumed. Within the eyes developmental and diabetic changes were combined. Because the cerebral and ocular malformations were comparably mild, the patient became unusually old, and visual function allowed at least good orientation. References were searched by medline and by the indices of known papers.

Expression and functional activity of heat shock proteins in human glioblastoma multiforme

OBJECTIVE

To assess the expression of heat shock proteins (HSP) in glioma cells in vitro and in vivo and to examine their role in resistance to apoptosis.

BACKGROUND

HSP are expressed in response to various forms of stress. Constitutive HSP expression may confer resistance to cytotoxic stimuli in human cancers.

METHODS

HSP expression was assessed by immunoblot analysis in glioma cells in vitro and by immunocytochemistry in human glioblastomas in vivo. Modulation of apoptosis by hyperthermia-mediated HSP induction was examined in glioma cell lines in vitro.

RESULTS

Immunoblot analysis revealed constitutive expression of HSP27, HSP72, HSP73, and HSP90 in all 12 human glioma cell lines. B-crystallin (alphaBC) was expressed in 3 of 12 cell lines. High levels of alphaBC and HSP72 correlated with drug resistance and high p53 levels in vitro. Transient hyperthermia (43 degrees C/2 hours) induced HSP27 and HSP72 expression but had no effect on the levels of alphaBC, HSP73, or HSP90. HSP induction provided no survival advantage against subsequent cytotoxic challenges, including cytotoxic cytokines and radiochemotherapy. Immunohistochemistry showed strong expression of all HSP in vivo. The comparative analysis of HSP27, alphaBC, HSP72, HSP73, and HSP90 expression in 24 paired samples of first resections and recurrences of human glioblastoma multiforme revealed no impact of HSP expression on response to adjuvant radiochemotherapy and no modulation of HSP expression by radiochemotherapy.

CONCLUSIONS

High constitutive, as opposed to inducible, expression of HSP may play a role in the primary resistance of human malignant gliomas to cytotoxic radiochemotherapy. Superinduction of HSP levels by hyperthermia in vitro provided no further survival advantage.

Quantitative assessment of second language learners' fluency by means of automatic speech recognition technology

To determine whether expert fluency ratings of read speech can be predicted on the basis of automatically calculated temporal measures of speech quality, an experiment was conducted with read speech of 20 native and 60 non-native speakers of Dutch. The speech material was scored for fluency by nine experts and was then analyzed by means of an automatic speech recognizer in terms of quantitative measures such as speech rate, articulation rate, number and length of pauses, number of dysfluencies, mean length of runs, and phonation/time ratio. The results show that expert ratings of fluency in read speech are reliable (Cronbach's alpha varies between 0.90 and 0.96) and that these ratings can be predicted on the basis of quantitative measures: for six automatic measures the magnitude of the correlations with the fluency scores varies between 0.81 and 0.93. Rate of speech appears to be the best predictor: correlations vary between 0.90 and 0.93. Two other important determinants of reading fluency are the rate at which speakers articulate the sounds and the number of pauses they make. Apparently, rate of speech is such a good predictor of perceived fluency because it incorporates these two aspects.

BCL-2 family protein expression in initial and recurrent glioblastomas: modulation by radiochemotherapy

OBJECTIVE

In vitro studies indicate a role of apoptosis regulatory proteins of the BCL-2 family in the resistance of glioblastoma multiforme to irradiation and chemotherapy. To date, no study has compared the expression of these proteins in initial and recurrent tumours. The differences of expression of BCL-2, BCL-X, BAX, and MCL-1 proteins of paired first resection and recurrence glioblastoma specimens were examined.

METHODS

Immunohistochemistry was performed in 37 cases of glioblastoma multiforme with paraffin embedded tissue from first resections and their recurrences in three treatment groups (15 radiochemotherapy, 15 irradiation, seven untreated). Ten high power fields were evaluated with an arbitrary score (< 5%=1, 5-50%=2, >50%=3), and cumulative scores for each antigen calculated.

RESULTS

In the whole group, we found a significant up regulation of antiapoptotic BCL-2 (median cumulative score of 15 in the primary, 19 at recurrence; p<0.0001 in the Wilcoxon test), BCLX (median scores 20 and 25, respectively, p<0.0001), and MCL-1 (median scores 11 and 14, p=0.0395), and a significant down regulation of proapoptotic BAX (median scores 14 and 11, p<0.0001). In the subgroups, these trends were also found. No association between protein expression and treatment regimen was found, although significant changes were restricted to the subgroups that received adjuvant chemotherapy. No significant correlation with clinical prognosis was detected with the Kaplan-Meier method.

CONCLUSIONS

In the development from initial to recurrent glioblastoma multiforme, the BCL-2 family rheostat shifts towards antiapoptotic adjustment in vivo. Importantly, the changes in BCL-2 family protein expression characterised here were also seen in the subgroup of patients who did not receive adjuvant radiotherapy or chemotherapy, suggesting that the changes of BCL-2 family protein expression result not only from radiochemotherapy but also reflect the natural course of disease.

Large cell anaplastic (KI-1) brain lymphoma of T-cell genotype

Three cerebral lesions were neuroradiologically detected in a 63-year-old man without evidence of an extracranial neoplasm. The biopsy specimen from one lesion showed a large cell anaplastic lymphoma (LCAL). Immunohistochemically, tumor cells were positive for CD3, CD30, CD45RO, and HLA-DR. The polymerase chain reaction (PCR) of the rearranged T-cell receptor beta chain genome (TcR beta) derived from paraffin sections showed monoclonality. This case shows that primary cerebral T-cell lymphomas genotypically correspond to nodal T-cell lymphomas, a correspondence that has previously been demonstrated for the more common brain lymphomas of B-cell lineage.