Staining, magnification, and algorithmic conditions for highly accurate cell detection and cell classification by deep learning

OBJECTIVES

Research into cytodiagnosis has seen an active exploration of cell detection and classification using deep learning models. We aimed to clarify the challenges of magnification, staining methods, and false positives in creating general purpose deep learning-based cytology models.

METHODS

Using 11 types of human cancer cell lines, we prepared Papanicolaou- and May-Grünwald-Giemsa (MGG)-stained specimens. We created deep learning models with different cell types, staining, and magnifications from each cell image using the You Only Look Once, version 8 (YOLOv8) algorithm. Detection and classification rates were calculated to compare the models.

RESULTS

The classification rates of all the created models were over 95.9%. The highest detection rates of the Papanicolaou and MGG models were 92.3% and 91.3%, respectively. The highest detection rates of the object detection and instance segmentation models, which were 11 cell types with Papanicolaou staining, were 94.6% and 91.7%, respectively.

CONCLUSIONS

We believe that the artificial intelligence technology of YOLOv8 has sufficient performance for applications in screening and cell classification in clinical settings. Conducting research to demonstrate the efficacy of YOLOv8 artificial intelligence technology on clinical specimens is crucial for overcoming the unique challenges associated with cytology.

Relationship between a deep learning model and liquid-based cytological processing techniques

OBJECTIVE

Artificial intelligence (AI)-based cytopathology studies conducted using deep learning have enabled cell detection and classification. Liquid-based cytology (LBC) has facilitated the standardisation of specimen preparation; however, cytomorphology varies according to the LBC processing technique used. In this study, we elucidated the relationship between two LBC techniques and cell detection and classification using a deep learning model.

METHODS

Cytological specimens were prepared using the ThinPrep and SurePath methods. The accuracy of cell detection and cell classification was examined using the one- and five-cell models, which were trained with one and five cell types, respectively.

RESULTS

When the same LBC processing techniques were used for the training and detection preparations, the cell detection and classification rates were high. The model trained on ThinPrep preparations was more accurate than that trained on SurePath. When the preparation types used for training and detection were different, the accuracy of cell detection and classification was significantly reduced (P < 0.01). The model trained on both ThinPrep and SurePath preparations exhibited slightly reduced cell detection and classification rates but was highly accurate.

CONCLUSIONS

For the two LBC processing techniques, cytomorphology varied according to cell type; this difference affects the accuracy of cell detection and classification by deep learning. Therefore, for highly accurate cell detection and classification using AI, the same processing technique must be used for both training and detection. Our assessment also suggests that a deep learning model should be constructed using specimens prepared via a variety of processing techniques to construct a globally applicable AI model.

Effect of liquid-based cytology fixing solution on immunocytochemistry: Efficacy of antigen retrieval in cytologic specimens

BACKGROUND

Immunocytochemistry (ICC) is an indispensable technique to improve diagnostic accuracy. ICC using liquid-based cytology (LBC)-fixed specimens has been reported. However, problems may arise if the samples are not fixed appropriately. We investigated the relationship between the LBC fixing solution and ICC and the usefulness of antigen retrieval (AR) in LBC specimens.

METHODS

Specimens were prepared from five types of LBC-fixed samples using cell lines and the SurePath™ method. ICC was performed using 13 antibodies and analyzed by counting the number of positive cells in the immunocytochemically stained specimens.

RESULTS

Insufficient reactivity was observed using ICC without heat-induced AR (HIAR) in nuclear antigens. The number of positive cells increased in ICC with HIAR. The percentage of positive cells was lower in CytoRich™ Blue samples for Ki-67 and in CytoRich™ Red and TACAS™ Ruby samples for estrogen receptor and p63 than in the other samples. For cytoplasmic antigens, the percentage of positive cells for no-HIAR treatment specimens was low in the three antibodies used. In cytokeratin 5/6, the number of positive cells increased in all LBC specimens with HIAR, and the percentage of positive cells in CytoRich™ Red and TACAS™ Ruby samples was significantly lower (p < .01). For cell membrane antigens, CytoRich™ Blue samples had a lower percentage of positive cells than the other LBC-fixed samples.

CONCLUSION

The combination of detected antigen, used cells, and fixing solution may have different effects on immunoreactivity. ICC using LBC specimens is a useful technique, but the staining conditions should be examined before performing ICC.

Effect of Specimen Processing Technique on Cell Detection and Classification by Artificial Intelligence

OBJECTIVES

Cytomorphology is known to differ depending on the processing technique, and these differences pose a problem for automated diagnosis using deep learning. We examined the as-yet unclarified relationship between cell detection or classification using artificial intelligence (AI) and the AutoSmear (Sakura Finetek Japan) and liquid-based cytology (LBC) processing techniques.

METHODS

The "You Only Look Once" (YOLO), version 5x, algorithm was trained on the AutoSmear and LBC preparations of 4 cell lines: lung cancer (LC), cervical cancer (CC), malignant pleural mesothelioma (MM), and esophageal cancer (EC). Detection and classification rates were used to evaluate the accuracy of cell detection.

RESULTS

When preparations of the same processing technique were used for training and detection in the 1-cell (1C) model, the AutoSmear model had a higher detection rate than the LBC model. When different processing techniques were used for training and detection, detection rates of LC and CC were significantly lower in the 4-cell (4C) model than in the 1C model, and those of MM and EC were approximately 10% lower in the 4C model.

CONCLUSIONS

In AI-based cell detection and classification, attention should be paid to cells whose morphologies change significantly depending on the processing technique, further suggesting the creation of a training model.

Immunocytochemical Reactivity Comparison between Formalin-Fixed and Liquid-Based Cytology-Fixed Specimens

INTRODUCTION

Liquid-based cytology (LBC)-fixed samples can be used for preparing multiple specimens of the same quality and for immunocytochemistry (ICC); however, LBC fixing solutions affect immunoreactivity. Therefore, in this study, we examined the effect of LBC fixing solutions on immunoreactivity.

METHODS

Samples were cell lines, and specimens were prepared from cell blocks of 10% neutral buffered formalin (NBF)-fixed samples and the four types of LBC-fixed samples: PreservCyt®, CytoRich™ Red, CytoRich™ Blue, and TACAS™ Ruby, which were post-fixed with NBF. ICC was performed using 24 different antibodies, and immunocytochemically stained specimens were analyzed for the percentage of positive cells.

RESULTS

Immunoreactivity differed according to the type of antigen detected. For nuclear antigens, the highest percentage of positive cells of Ki-67, WT-1, ER, and p63 was observed in the NBF-fixed samples, and the highest percentage of positive cells of p53, TTF-1, and PgR was observed in the TACAS™ Ruby samples. For cytoplasmic antigens, the percentage of positive cells of CK5/6, Vimentin, and IMP3 in LBC-fixed samples was higher than or similar to that in NBF-fixed samples. The percentage of positive cells of CEA was significantly lower in CytoRich™ Red and CytoRich™ Blue samples than in the NBF-fixed sample (p < 0.01). Among the cell membrane antigens, the percentage of positive cells of Ber-EP4, CD10, and D2-40 was the highest in NBF-fixed samples and significantly lower in CytoRich™ Red and CytoRich™ Blue samples than that in NBF-fixed samples (p < 0.01). The NBF-fixed and LBC-fixed samples showed no significant differences in the percentage of positive cells of CA125 and EMA.

DISCUSSION/CONCLUSION

ICC using LBC-fixed samples showed the same immunoreactivity as NBF-fixed samples when performed on cell block specimens post-fixed with NBF. The percentage of positive cells increased or decreased based on the type of fixing solution depending on the amount of antigen in the cells. Further, the detection rate of ICC with LBC-fixed samples varied according to the type of antibody and the amount of antigen in the cells. Therefore, we propose that ICC using LBC-fixed samples, including detection methods, should be carefully performed.

Relationship between Liquid-Based Cytology Preservative Solutions and Artificial Intelligence: Liquid-Based Cytology Specimen Cell Detection Using YOLOv5 Deep Convolutional Neural Network

INTRODUCTION

Deep learning is a subset of machine learning that has contributed to significant changes in feature extraction and image classification and is being actively researched and developed in the field of cytopathology. Liquid-based cytology (LBC) enables standardized cytological preparation and is also applied to artificial intelligence (AI) research, but cytological features differ depending on the LBC preservative solution types. In this study, the relationship between cell detection by AI and the type of preservative solution used was examined.

METHODS

The specimens were prepared from five preservative solutions of LBC and stained using the Papanicolaou method. The YOLOv5 deep convolutional neural network algorithm was used to create a deep learning model for each specimen, and a BRCPT model from five specimens was also created. Each model was compared to the specimen types used for detection.

RESULTS

Among the six models, a difference in the detection rate of approximately 25% was observed depending on the detected specimen, and within specimens, a difference in the detection rate of approximately 20% was observed depending on the model. The BRCPT model had little variation in the detection rate depending on the type of the detected specimen.

CONCLUSIONS

The same cells were treated with different preservative solutions, the cytologic features were different, and AI clarified the difference in cytologic features depending on the type of solution. The type of preservative solution used for training and detection had an extreme influence on cell detection using AI. Although the accuracy of the deep learning model is important, it is necessary to understand that cell morphology differs depending on the type of preservative solution, which is a factor affecting the detection rate of AI.

Pharmacological inhibition of the lipid phosphatase PTEN ameliorates heart damage and adipose tissue inflammation in stressed rats with metabolic syndrome

Phosphatidylinositol 3-kinase (PI3K) signaling promotes the differentiation and proliferation of regulatory B (Breg) cells, and the lipid phosphatase phosphatase and tensin homolog deleted on chromosome 10 (PTEN) antagonizes the PI3K-Akt signaling pathway. We previously demonstrated that cardiac Akt activity is increased and that restraint stress exacerbates hypertension and both heart and adipose tissue (AT) inflammation in DS/obese rats, an animal model of metabolic syndrome (MetS). We here examined the effects of restraint stress and pharmacological inhibition of PTEN on heart and AT pathology in such rats. Nine-week-old animals were treated with the PTEN inhibitor bisperoxovanadium-pic [bpV(pic)] or vehicle in the absence or presence of restraint stress for 4 weeks. BpV(pic) treatment had no effect on body weight or fat mass but attenuated hypertension in DS/obese rats subjected to restraint stress. BpV(pic) ameliorated left ventricular (LV) inflammation, fibrosis, and diastolic dysfunction as well as AT inflammation in the stressed rats. Restraint stress reduced myocardial capillary density, and this effect was prevented by bpV(pic). In addition, bpV(pic) increased the proportions of Breg and B-1 cells as well as reduced those of CD8+ T and B-2 cells in AT of stressed rats. Our results indicate that inhibition of PTEN by bpV(pic) alleviated heart and AT inflammation in stressed rats with MetS. These positive effects of bpV(pic) are likely due, at least in part, to a reduction in blood pressure, an increase in myocardial capillary formation, and an altered distribution of immune cells in fat tissue that result from the activation of PI3K-Akt signaling.

Characterizing the Effect of Processing Technique and Solution Type on Cytomorphology Using Liquid-Based Cytology

INTRODUCTION

Liquid-based cytology (LBC) is increasingly used for nongynecologic applications. However, the cytological preparation of LBC specimens is influenced by the processing technique and the preservative used. In this study, the influence of the processing techniques and preservatives on cell morphology was examined mathematically and statistically.

METHODS

Cytological specimens were prepared using the ThinPrep (TP), SurePath (SP), and AutoSmear methods, with 5 different preservative solutions. The cytoplasmic and nuclear areas of Papanicolaou-stained specimens were measured for all samples.

RESULTS

The cytoplasmic and nuclear areas were smaller in cells prepared using the 2 LBC methods, compared to that prepared using the AutoSmear method, irrespective of the preservative used. The cytoplasmic and nuclear areas of cells prepared using the SP method were smaller than those of cells prepared using the TP method, irrespective of the preservative used. There were fewer differences among the cytoplasmic areas of cells prepared with different preservative solutions using the TP method; however, the cytoplasmic areas of cells prepared using the SP method changed with the preservative solution used.

CONCLUSIONS

The most significant difference affecting the cytoplasmic and nuclear areas was the processing technique. The TP method increased the cytoplasmic and nuclear areas, while the methanol-based PreservCyt solution enabled the highest enlargement of the cell. LBC is a superior preparation technique for standardization of the specimens. Our results offer a better understanding of methods suitable for specimen preparation for developing precision AI-based diagnosis in cytology.

U-shape rotating anti-cathode compact X-ray generator: 20 times stronger than the commercially available X-ray source

A new type of U-shape anti-cathode X-ray generator in which the inner surface of a cylindrical target is irradiated by an electron beam has been made by modifying a conventional rotating anti-cathode X-ray generator whose brightness in the catalog is 12 kW mm(-2). The target material (Cu), target radius (50 mm) and rotating speed (6,000 r.p.m.) were not changed in this modification. A brightness of 52 kW mm(-2) was obtained by this U-shape-type X-ray generator. This means that the brightness of the new type is 4.3 times greater than that of the old unmodified one. Furthermore, the new-type X-ray generator yielded a brightness of 129 kW mm(-2) by adding a carbon coating on the Cu target. This means an overall increase of brightness of ten times. The original generator has the highest brightness in the generators of the same class (having a radius of 50 mm and rotation speed of 6,000 r.p.m.). Observations showed that Cu Kα counts at vertical incidence of the electron beam onto the surface of the new target, which is initially optically smooth, decrease as the surface is roughened by a severe thermal stress caused by strong electron beam exposure. Further observation reveals, however, that oblique incidence of the electron beam onto the roughened surface drastically increased the X-ray output and amounts to twice as much as that from a smooth surface at vertical incidence. Thus, at the present stage, an overall increase of brightness has been realised at a level 20 times stronger than that of the original commercially offered X-ray generator that we modified.

Highly bright X-ray generator using heat of fusion with a specially designed rotating anticathode

A new type of rotating anticathode X-ray generator has been developed, in which the electron beam irradiates the inner surface of a U-shaped anticathode (Cu). A high-flux electron beam is focused on the inner surface by optimizing the shape of the bending magnet. The power of the electron beam can be increased to the point at which the irradiated part of the inner surface is melted, because a strong centrifugal force fixes the melted part on the inner surface. When the irradiated part is melted, a large amount of energy is stored as the heat of fusion, resulting in emission of X-rays 4.3 times more brilliant than can be attained by a conventional rotating anticathode. Oscillating translation of the irradiated position on the inner surface during use is expected to be very advantageous for extending the target life. A carbon film coating on the inner surface is considered to suppress evaporation of the target metal and will be an important technique in further realization of highly bright X-ray generation.

Conceptual design of novel IP-conveyor-belt Weissenberg-mode data-collection system with multi-readers for macromolecular crystallography. A comparison between Galaxy and Super Galaxy

Galaxy is a Weissenberg-type high-speed high-resolution and highly accurate fully automatic data-collection system using two cylindrical IP-cassettes each with a radius of 400 mm and a width of 450 mm. It was originally developed for static three-dimensional analysis using X-ray diffraction and was installed on bending-magnet beamline BL6C at the Photon Factory. It was found, however, that Galaxy was also very useful for time-resolved protein crystallography on a time scale of minutes. This has prompted us to design a new IP-conveyor-belt Weissenberg-mode data-collection system called Super Galaxy for time-resolved crystallography with improved time and crystallographic resolution over that achievable with Galaxy. Super Galaxy was designed with a half-cylinder-shaped cassette with a radius of 420 mm and a width of 690 mm. Using 1.0 A incident X-rays, these dimensions correspond to a maximum resolutions of 0.71 A in the vertical direction and 1.58 A in the horizontal. Upper and lower screens can be used to set the frame size of the recorded image. This function is useful not only to reduce the frame exchange time but also to save disk space on the data server. The use of an IP-conveyor-belt and many IP-readers make Super Galaxy well suited for time-resolved, monochromatic X-ray crystallography at a very intense third-generation SR beamline. Here, Galaxy and a conceptual design for Super Galaxy are described, and their suitability for use as data-collection systems for macromolecular time-resolved monochromatic X-ray crystallography are compared.

Large-format imaging plate and weissenberg camera for accurate protein crystallographic data collection using synchrotron radiation

Off-line and on-line protein data-collection systems using an imaging plate as a detector are described and their components reported. The off-line scanner IPR4080 was developed for a large-format imaging plate ;BASIII' of dimensions 400 x 400 mm and 400 x 800 mm. The characteristics of this scanner are a dynamic range of 10(5) photons pixel(-1), low background noise and high sensitivity. A means of reducing electronic noise and a method for finding the origin of the noise are discussed in detail. A dedicated screenless Weissenberg camera matching IPR4080 with synchrotron radiation was developed and installed on beamline BL6B at the Photon Factory. This camera can attach one or two sheets of 400 x 800 mm large-format imaging plate inside the film cassette by evacuation. The positional reproducibility of the imaging plate on the cassette is so good that the data can be processed by batch job. Data of 93% completeness up to 1.6 A resolution were collected on a single axis rotation and the value of R(merge) becomes 4% from a tetragonal lysozyme crystal using a set of two imaging-plate sheets. Comparing two types of imaging plates, the signal-to-noise ratio of the ST-VIP-type imaging plate is 25% better than that of the BASIII-type imaging plate for protein data collection using 1.0 and 0.7 A X-rays. A new on-line protein data-collection system with imaging plates is specially designed to use synchrotron radiation X-rays at maximum efficiency.

Time-resolved protein crystallography with large-angle oscillations: an application of a protein data-collection system using the Weissenberg technique and a large-format imaging plate

A diffraction-intensity data-collection system with synchrotron radiation X-rays utilizing the screenless Weissenberg technique and incorporating a large-format imaging plate is one of the most suitable apparatus for time-resolved protein crystallography with larger angle oscillations than hitherto described. The time resolution and data quality of the system have been tested using a tetragonal lysozyme crystal as a test sample in a flow-cell experiment at the bending-magnet beamline 18B at the Photon Factory, and a time resolution of 15 min is confirmed.

A preliminary description of the crystal structure of gamma-glutamyltranspeptidase from E. coli K-12

The structure of GGT [EC 2.3.2.2] from E. coli K-12 was studied at 3 A resolution by X-ray crystallography. Initial protein phases were calculated using two kinds of Pb2+ derivatives. The phases were refined by non-crystallographic 2-fold symmetry electron density averaging combined with solvent flattening and histogram matching. The GGT molecule has overall dimensions of 60 x 50 x 40 A. There are two antiparallel beta-pleated sheets consisting of 6 and 7 beta-strands. The two beta-sheets form a wall-like structure. Twelve short alpha-helices were detected, of which the maximum length appears to be four helix turns.

Crystallization and preliminary X-ray analysis of gamma-glutamyltranspeptidase from Escherichia coli K-12

gamma-Glutamyltranspeptidase (EC 2.3.2.2) from Escherichia coli K-12 has been purified and crystallized by means of vapor diffusion in hanging drops. Two kinds of crystals on cell dimensions were found for X-ray diffraction analysis, one from ammonium sulfate and the other from polyethylene glycol 6000 as precipitants. The crystals of the orthorhombic form grown in the presence of 15% polyethylene glycol and 20 mM sodium acetate buffer were chosen for further analysis. The crystals belonged to space group P2(1)2(1)2(1), with cell dimensions of a = 128.1, b = 129.9 and c = 79.2 A, and two molecules constitute an asymmetric unit. These crystals diffracted to 2.0 A resolution and were suitable for X-ray crystallographic studies.

Electrocardiogram studies in llamas

Electrocardiograms (ECG) of 3 captive llamas (Lama glama) were recorded by a telemeter at a farm in Japan. The pattern of the ECGs was similar to the ruminant pattern in the AB lead position. QRS and T-waves were discordant in polarity except in one llama, where the polarity of the T wave changed according to the HR. In the quiet state, the HR varied between 60-80/min depending on the nervousness of the llama. After running and after being held, the HR increased to more than 100/min. During the recordings there were some variations of the HR which could be due to respiratory arrhythmia. Second-degree AV block and supraventricular premature complexes were also found in two llamas.

Crystal structure analysis of omega-amino acid:pyruvate aminotransferase with a newly developed Weissenberg camera and an imaging plate using synchrotron radiation

The three-dimensional structure of omega-amino acid:pyruvate aminotransferase from Pseudomonas sp. F-126, an isologous alpha 4 tetramer containing pyridoxal 5'-phosphate (PLP) as a cofactor, has been determined at 2.0 A resolution. The diffraction data were collected with a newly developed Weissenberg camera with a Fuji Imaging Plate, using synchrotron radiation. The mean figure-of-merit was 0.57. The subunit is rich in secondary structure and comprises two domains. PLP is located in the large domain. The high homology in the secondary structure between this enzyme and aspartate aminotransferase strongly indicates that these two types of enzymes have evolved from a common ancestor.

X-ray diffraction photographs of chicken gizzard G-actin.DNase I complex crystals taken with synchrotron radiation

X-ray diffraction photographs of a chicken gizzard G-actin.DNase I complex crystal have been recorded using the synchrotron radiation beam emitted by the Synchrotron Radiation Source at Daresbury and the Photon Factory at Tsukuba. The resolution limit was extended to 2.4 A and the exposure time was reduced approximately by a factor of 10, when data recorded at the Photon Factory, were compared with those recorded with a conventional rotating-anode source. Using a newly designed Weissenberg camera equipped with a multi-layer line screen, the diffraction data in a 36 degrees oscillation range were recorded on a single film up to 3.5 A resolution.

Crystallographic studies of the chicken gizzard G-actin X DNase I complex at 5A resolution

The structure of the chicken gizzard G-actin X DNase I complex has been determined at 5 A resolution by an X-ray diffraction method. Protein phases were computed by the multiple isomorphous replacement method using four heavy atom derivatives. The mean figure of merit was 0.65. Dimensions of the three molecular species, the complex, G-actin and DNase I, were determined based on the "cypress wood" models derived from the electron density map. The natures of the heavy atom binding sites are discussed in relation to the distinction between the two component molecules. The pattern of successive contacts between actin molecules observed in the present crystal seems unrelated to that found in F-actin.

Power spectral analysis of auditory brain stem responses to pure tone stimuli

Power spectral analysis and digital filtration of the auditory brain stem responses to tone burst stimuli were carried out in three adult subjects with normal hearing. Three dominant peaks (peaks A. B, C), ranging in frequency 50-150, 500-600, and 1000-1100 Hz, were observed in the spectrograms of the responses to 4 kHz at 80 dB SL. Power of higher frequency components (peaks B and C) decreased markedly or disappeared completely with decreased intensity as well as decreased frequency of stimulus, while peak A around 100 Hz consistently appeared regardless of stimulus intensity or frequency. Main spectral components of the individual waves in the ABR were determined as follows: peak A for the slow positive deflection in the response, peak B for waves VI and VII, peak C for waves II and IV, and peaks B and C for waves I, III and V.

Power spectral analysis of auditory evoked response

Five normal hearing adults were examined for power spectral analysis of AER to 1000-Hz pure tone. Major frequency components of responses to stimuli at 50 dB hearing level exist at 1 to 14 c/s. The power spectra of responses to stimuli at 10 dB hearing level are greater than those of the SBA in the frequency components between 3 and 9 c/s; furthermore, the total power of 3 to 9 c/s indicates a clear difference between the 2. Incidentally, at 512-msec analysis time, the total power of 3 to 9 c/s of the AER are significantly greater than that at 1024-msec analysis time, whereas those of the SBA show little discrepancy at the 2; thus at 512-msec analysis time, the total power of 4 to 9 c/s yields a clear difference between the AER and the SBA responses averaged over 50 stimuli even at 10 dB hearing level.

Crystallization and preliminary crystallographic data of chicken gizzard G-actin . DNase I complex and Physarum G-actin . DNase I complex

Smooth muscle G-actin from chicken gizzard and Physarum plasmodium G-actin both interact with DNase I and form 1 : 1 complexes. These complexes were crystallized by using polyethylene glycol 6000 as a precipitant. Both crystals belong to the same orthorhombic space group P2(1)2(1)2(1). The cell dimensions of chicken gizzard G-actin.DNase I complex are a=42.00 +/- 0.07 A, b=225.3 +/- 0.4 A, and c=77.4 +/- 0.1 A, while those of Physarum G-actin.DNase I complex are a=42 A, b=221 A, and c=77 A.

Modification and application of the computerized automatic audiometer

Recently, medical information processing systems applying computers have achieved marked advances in parallel with the development of highly accurate automatic measuring instruments. The authors already in 1974 developed and reported a computerized automatic audiometer containing a micro-computer which stored a program for audiometric procedures. Subsequently the authors have developed a compact, multi-purpose audiometric instrument which executes the standard air-bone conduction audiometry, with the exclusion of masking, automatically with a built-in micro-computer. An outline of the instrument is as follows: 1. The instrument is composed of a memory, a control unit, programs for data processing and measurement, a character display for instructing subjects and an audiogram display. It can be operated manually in the same way as a conventional audiometer. 2. Test errors and any abnormally large differences between the responses of the two ears can be detected automatically, and the tests can then be re-executed manually after finishing the automatic procedure for correction of the data. 3. The instrument is provided with output terminals for connection to an external computer or other peripheral equipment. It is designed to be compact and light weight.

Development of an audiometry data processing system

The development was reported of an electronic data processing system of computerized automatic audiometry and of Bekesy audiometry. This system using a microcomputer has the following features: (1) The large amount of data obtained through (a) automatic audiometry, (b) Bekesy audiometry (measurement at fixed or continuous frequency plus test practice, and (c) the Temporal Tone-Decay test can be stored on-line in real time, and are processed and displayed off-line at any time. (2) When retrieved, the output format of these audiometric data is of the same pattern as the conventional audiogram. (3) The output pattern on the CRT graphic display can be hard-copied whenever desired. (4) The system can be operated manually in the same way as the conventional method. (5) The SISI and DL tests can be executed manually and the resultant data keyed into computer storage. And (6) The operation from the data input to the retrieval output is performed interpretively through the CRT display, so that anyone even without special computer or audiological training can operate it at any time.