Search PubMedSearch

SEARCH · Search PubMed

Results for “Compression Algorithms”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3Linked to original sources

Application of artificial neural network to computer-aided diagnosis of coronary artery disease in myocardial SPECT bull's-eye images.

We have developed a computerized system that can aid in the radiologist's diagnosis in the detection and classification of coronary artery diseases. The technique employs a neural network to analyze 201Tl myocardial SPECT bull's-eye images. This multi-layer feed-forward neural network with a backpropagation algorithm has 256 input units (pattern: compressed 16 x 16-matrix images), 5-140 units in a single hidden layer, and eight output units (diagnosis: one normal and seven different types of abnormalities). The neural network was taught using pairs of training (learning) input data (bull's-eye "EXTENT" image) and desired output data ("correct" diagnosis). The effects of the numbers of hidden units and learning iterations in the network on the recognition performance were examined. In our initial stage, the results show that the recognition performance of the neural network is better than that of the radiology resident but worse than that of the experienced radiologist. Our study also demonstrates that the result produced in the neural network depends on the variety of the training examples used. The preliminary study suggests that the neural network approach is useful for the computer-aided diagnosis of coronary artery diseases in myocardial SPECT bull's-eye images.

Coronary Disease

Special problems in the older cancer patient: spinal cord compression and pleural effusions.

Spinal cord compression due to epidural metastases, and malignant pleural effusions are devastating complications of advanced cancer that can destroy the quality of a life that is already limited in quantity. Neither need do so. A diagnostic algorithm has been developed for each of these complications. Early diagnosis of cord compression can be accomplished by prompt myelography or magnetic resonance imaging of the affected area of the spine. A malignant pleural effusion, even one that initially appears cytologically negative, can be promptly diagnosed. Radiation and steroids are optimal therapy for most patients with cord compression, and few require surgery. Intrapleural bleomycin appears to be the most effective agent for pleurodesis.

Aging

MRI of intracranial neurovascular compression.

Twelve patients with clinical indications of intracranial neurovascular compression (6, hemifacial spasm; 4, trigeminal neuralgia; 2, vertigo and tinnitus) were studied by MRI. Axial 1 mm thick slices were obtained in a 1.5 T unit using a three-dimensional (3D) fast low angle shot data set of 32 contiguous slices. The data were post-processed with multiplanar reconstruction algorithms to obtain oblique sagittal (along the long axis of a nerve) images or coronal images with a slice thickness of 0.8 mm. The MR angiographic images were also obtained from the same 3D data set with maximum intensity projection algorithms. The MR studies showed that 9 of 12 patients had neurovascular compression caused by one or two arteries. These findings were verified at surgery. In two of the remaining three patients, MRI failed to delineate the affected nerves compressed by a vein in one, and a previously applied prosthesis in the other. In the last patient no neurovascular compression was found by MRI or at surgery.

Aged

Digital imaging and picture archiving and communication systems.

Great progress has been made in digital imaging of the chest. Most studies are dealing with computed radiography. Chest radiography in the intensive care unit may, in most cases, be performed using computed radiography. However, subtle pulmonary interstitial disease can be demonstrated less confidently using computed radiography. Significantly better detection of calcified lung nodules can be obtained by using simplified single-exposure dual-energy technique that uses storage phosphor. The wide latitude of computed radiography permits images of high quality in areas other than chest radiography. Encouraging results are presented especially in the diagnostic evaluation of scoliosis and other musculoskeletal abnormalities. An important technical innovation in digital radiography is an improved method for single-exposure dual-energy digital imaging using prefiltration with gadolinium, a cassette consisting of four photostimulable phosphor plates, spatially dependent scatter and beam hardening corrections, and noise reduction algorithm. Other groups tested algorithms for enhancement of digital images that allowed significant data compression. The implementation of picture archiving and communication systems (PACS) is inevitable; the question concerning PACS implementation is not why, but when. A comparison of the cost-effectiveness of PACS with conventional film archiving and communication systems shows that PACS should provide indirect savings when regarding the hidden costs of conventional systems. Much more experience will be needed before there is general agreement on the best design for the radiologist's workstation. Teleradiology should contribute to radiologic consultation for remote locations, because it improves the efficacy of management of patients in such locations.

Child

JPEG compression for PACS.

In the medical field, especially in diagnostic radiology, there still remains controversy over how and whether or not to compress X-ray images for storage and transmission. The joint Photographic Expert Group (JPEG) standard which has recently been agreed upon is the very attractive technique to archive and to transport images in medical fields. This technique is based on 'lossy' compression of images, which can handle not only X-ray images but also full-colored images, and is suitable for introduction into picture archiving and communicating systems (PACS). The images can be handled after compression as quite small clusters of data. For example, a single 2000 x 2000 x 12 bits chest X-ray image which is an 8 Mbyte image compressed at a 10:1 ratio could retain virtually all the visible quality of the original version, and would take 100 s to transmit at 64 kbits/s using the Integrated Services Digital Network (ISDN, 800 kbytes compressed file, 8 kbytes/s transmission theoretically). An important factor in the design of this technique is that this format relies on no specific hardware or software if using JFIF (JPEG File interchange Format). Soon this algorithm will be able to run on any workstation or on any PC in the world. Highly compressed images may be unsuitable for diagnostic purposes. However, they may be sufficient for reference images which will be needed in clinical fields.

Algorithms

[The computer-assisted processing of CT data in planning endosseous implantation interventions].

The success of dental implantologic measures depends considerably on the correct positioning of the endosseous implants. A sufficient amount of bony substance is mandatory to provide overall covering of the implant with at least 1 mm of cortical bone. By the aid of a custom-developed software system, computed tomography image data can be handled on a conventional MS-DOS personal computer. The user is enabled to simulate size and positioning of the implant by interactive graphic animation. Quick and exact presentation is achieved by optimizing the computation algorithms. Storage needs are minimized by a segmentation procedure and additional compression of the image. Integrated patient data management and user friendly self-documenting program surfaces provide easy application.

Computer Graphics

A new method for aligning histological serial sections for three-dimensional reconstruction.

A new method for aligning histological serial sections for three-dimensional reconstruction by computer is introduced. This method is particularly suited for embedding structures in celloidine or paraplast. Prior to sectioning, at least three reference markers are affixed to the preparation in a direction perpendicular to the section plane, for easy identification on the finished sections. An algorithm translates and rotates the serial sections until all the reference markers are congruent. The main advantage of this method, however, is its capability to eliminate deformations caused by compression or stretching during sectioning by the determination of correction factors.

Algorithms

Efficient storage of urodynamic signals by computer: application of FAN adaptive sampling.

Digital storage of urodynamic signals such as detrusor pressure and flowrate at a sufficiently high sampling rate (10 samples per second) to allow subsequent analysis requires considerable computer memory. A procedure for compressing these data by deleting redundant samples (the fan method of adaptive sampling) was tested. The method allows a flexible adaptation to specific hardware and a compromise between storage requirements and accuracy. In this study the number of samples required for adequate reconstruction of the detrusor pressure signal could be varied from 80% to 4% of the original number of samples by varying the average difference between reconstructed and original signal from 0.01 to 2 cm H2O. Fast components of the measurements (for example cough peaks) which were lost if a lower sampling rate or averaging was used to obtain equally low storage requirements were unaffected by this compression technique.

Algorithms

Visual disturbances of migraine.

Migraine, a clinical syndrome of unknown etiology, is a common cause of a variety of visual disturbances. This review describes the visual alterations associated with migraine syndromes of particular interest to the ophthalmologist; acephalgic, ocular, and ophthalmoplegic. Several current theories of migraine pathophysiology are discussed. Migrainous episodes are common and must be differentiated from neurologic dysfunction due to ischemia, inflammation, seizure, and compression. The differentiating characteristics of these conditions as well as a diagnostic algorithm are presented.

Adolescent

Two methods for medical image compression.

Depending on the purpose for which a medical image is taken, some areas of it contain more clinical information than others. These areas are usually considered as regions of interest (ROI). The area outside the ROI is either the background or contains some clinical information with relative importance. In this paper we introduce two preprocessing and error-free compression routines. The first preprocessing routine, if applied to X-ray images, could result in a reconstructed image with a controllable amount of deterioration in its various areas. The second preprocessing routine is applied to X-ray CT, MR and radionuclide images prior to the compression. Its aim is to automatically detect the ROI and neglect the background. The background is replaced by an optional gray level when these images are being reconstructed. These routines were applied to a variety of medical images and relatively good compression ratios were obtained.

Algorithms

[A computerized manual for assessing the dosage absorbed by breast tissue in mammography].

The computerized translation of "Handbook of glandular tissue doses in mammography" (HHS publication FDA 85-8239) is presented. It is registered on floppy disk and suitable for use in MS-DOS on IBM personal computer, Olivetti or compatible. The manual collects data published by many authors and suggests a method to estimate the dose released to the glandular tissue (considered at risk for breast cancer) in the most commonly used conditions for mammography. Besides translation manual, the program holds an algorithm developed by the authors, which allows the dose to the glandular tissue to be evaluated, given the following data: projection: craniocaudal or mediolateral; compression level: firm or moderate; breast size: small, medium or large; breast composition: percentage of glandular tissue content (by weight); X-ray beam quality: HVL mm Al. The algorithm is obtained from a bidimensional fit of the original data and is based on the terms contained in the manual. The program flow is guided by menus that make the procedure suitable for inexperienced operators too, both for consulting and printing and for computing program.

Algorithms

The landscape of pruning for large language models: A systematic review and unified taxonomy.

Confronting the inherent tension between the exceptional capabilities and the immense computational costs of Large Language Models (LLMs), pruning has become a crucial technique for achieving efficient deployment. However, a systematic analytical framework dedicated specifically to LLM pruning remains absent. In this paper, we aim to bridge this gap. We first elucidate the theoretical foundations that underpin the effectiveness of pruning, namely overparameterization and redundancy, and then propose a multidimensional taxonomy that organizes existing approaches along the axes of granularity, timing, and criteria. Building upon this unified perspective, we further analyze performance recovery mechanisms and the broader evaluation ecosystem, while also exploring forward-looking challenges such as interpretability, automation, and hardware-algorithm co-design. Through this comprehensive synthesis, we seek to provide an integrated and coherent analytical lens for advancing both research and practice in LLM pruning.

Large Language Models

Automated electrocardiogram analysis: the state of the art.

The overwhelming number of electrocardiograms (ECGs) now recorded routinely has prompted the development of computer analysis which in turn has benefited electrocardiography with important technological advances. All automated ECG analysis systems adopt a similar approach: a measurement program and a program that interprets the clinical significance of these measurements along with a rhythm analysis algorithm. Measurement, selection and classification of parameters vary according to the program used. Data compression is applied to the signal to reduce processing time and allow long-term storage. Diagnostic accuracy, however, is not greatly improved over that of experienced cardiologists. Programs studied using a validated data bank provided by an international group of cardiologists show a variability not only in parameter measurement but also in diagnostic statement and in the way in which such statements are expressed. Recommendations for measurement standards have been made to fulfil the need for exchange of diagnostic criteria. No recommendations concerning the selection of parameters have been proposed, and so new parameters or combinations of parameters can be introduced with the ultimate aim of increased diagnostic performance.

Diagnosis, Computer-Assisted

Radiology of acute spinal trauma.

Radiologic diagnostic modalities utilized to evaluate acute spine trauma include plain films, tomography, computerized tomography (CT), myelography and CT-myelography, and magnetic resonance imaging. The initial evaluation is aimed at delineating the nature and extent of both boney and soft tissue injuries, with particular attention directed toward differentiating intrinsic cord damage from extrinsic spinal cord or nerve root compression that may respond to surgical intervention. The selection of the optimal radiologic diagnostic algorithm requires familiarization with the techniques and limitations of each diagnostic procedure.

Algorithms

Osteoporosis: the structural and reparative consequences for the skeleton.

We have discussed the biomechanical and biomaterial properties of bone with emphasis on the microenvironment. The microdensity of bone tissues determines the resistance to compressive forces (strength = density2). The structural orientation of bone determines the ability of bone to withstand torsion and bending forces. Highly remodeled bone is brittle and fractures easily because of the multiple reversal planes within the bone plates. Conversely, augmentation of bone on the periphery by modeling leads to increased material microstrength and structural macrostrength. Consequently, appositional modeling augments whereas remodeling diminishes bone strength. Hip fractures, spinal fractures, Colles' fractures, and pelvic fractures have been discussed in terms of their pathophysiology and treatment. An algorithm (Fig. 4-7) has been presented for the differential diagnosis of patients with recently discovered spinal compression fractures. Lastly, a discussion of exercise demonstrated its beneficial role, especially in the retention of bone mass.

Biomechanical Phenomena

Storage volume of computer tomography images can be reduced by a factor of five.

The object of this study is the data compression of CT scans of the brain. A circle is first established which contains all pixels inside the head while minimizing the insignificant area, after which the portion outside the circle so obtained is eliminated. The data are then truncated at CT values of +/- 127 and the differences of two consecutive pixel values are converted into variable length codes. The coding algorithms employed are the well-known Huffman code and our packet coding, both of which achieved a data compression ratio of 20%.

Biometry

An algorithm to manage variable-length records for highly portable clinical data base systems.

An algorithm to archive patient data at free size in disk storage is presented. A record, assumed to be a character string such as an ASCII-coded text, is compressed and divided into fixed-length blocks. One block consists of a data field and a pointer field, and the blocks comprising a record are chained with pointers forwardly. A head pointer of each record is sequentially saved on a separate file. The data compression is performed as follows: if the same character code appears more than twice in succession, we count the number of the repetitions and save it with initial two characters. The algorithms for fetching, re-saving, and purging a record are also presented. These were implemented in FORTRAN77 and tested for performance using a practical patient data file. As the algorithm allows highly flexible record manipulation and can easily be implemented in conventional programming languages, it will make a useful tool for constructing a portable data base management system.

Evaluation Studies as Topic