Machine-readable Code Research Articles

A community-endorsed open-source lexicon for contrast agent-based perfusion MRI: A consensus guidelines report from the ISMRM Open Science Initiative for Perfusion Imaging (OSIPI).

This manuscript describes the ISMRM OSIPI (Open Science Initiative for Perfusion Imaging) lexicon for dynamic contrast-enhanced and dynamic susceptibility-contrast MRI. The lexicon was developed by Taskforce 4.2 of OSIPI to provide standardized definitions of commonly used quantities, models, and analysis processes with the aim of reducing reporting variability. The taskforce was established in February 2020 and consists of medical physicists, engineers, clinicians, data and computer scientists, and DICOM (Digital Imaging and Communications in Medicine) standard experts. Members of the taskforce collaborated via a slack channel and quarterly virtual meetings. Members participated by defining lexicon items and reporting formats that were reviewed by at least two other members of the taskforce. Version 1.0.0 of the lexicon was subject to open review from the wider perfusion imaging community between January and March 2022, and endorsed by the Perfusion Study Group of the ISMRM in the summer of 2022. The initial scope of the lexicon was set by the taskforce and defined such that it contained a basic set of quantities, processes, and models to enable users to report an end-to-end analysis pipeline including kinetic model fitting. We also provide guidance on how to easily incorporate lexicon items and definitions into free-text descriptions (e.g., in manuscripts and other documentation) and introduce an XML-based pipeline encoding format to encode analyses using lexicon definitions in standardized and extensible machine-readable code. The lexicon is designed to be open-source and extendable, enabling ongoing expansion of its content. We hope that widespread adoption of lexicon terminology and reporting formats described herein will increase reproducibility within the field.

Making humanoid robots teaching assistants by using natural language processing (NLP) cloud-based services

Humanoid robots have a substantial potential to serve as teaching and social assistants. However, the expectations of the children from robots to interact like humans are huge. This study presents a general model for understanding the natural language in human-robot interaction by applying Generative Pre-trained Transformer (GPT) language models as a service in the Internet of Things. Thus, the physical presence of the robot can help in fine-tuning the GPT model by prompts derived from the environmental context and subsequent robot actions for embodiment understanding of the GPT outputs. The model uses web or cloud services for Natural Language Processing (NLP) to produce and play human-like text, question answering or text generation. Verbal questions are processed either via a local speech recognition software or via a Speech-to-Text (STT) cloud service. The converted question into machine-readable code is sent to one of the GPT language models with zero- or few-shot learning prompts. GPT-J model has been tested and deployed either in the web or cloud with options for varying the parameters for controlling the haphazardness of the generated text. The robot produces human-like text by using Text-to-Speech (TTS) cloud services that convert the response into audio format rendered on the robot to be played. Useful requirements how the model to be used in order to be feasible were determined based on the evaluation of the outputs given from the different NLP and GPT-J web or cloud-services. We designed and implemented the model in order to be used by a humanoid NAO-type robot in the speech language therapy practice, however it can be used for other open-source and programmable robots and in different contexts.

Towards Facial Biometrics for ID Document Validation in Mobile Devices

Various modern security systems follow a tendency to simplify the usage of the existing biometric recognition solutions and embed them into ubiquitous portable devices. In this work, we continue the investigation and development of our method for securing identification documents. The original facial biometric template, which is extracted from the trusted frontal face image, is stored on the identification document in a secured personalized machine-readable code. Such document is protected from face photo manipulation and may be validated with an offline mobile application. We apply automatic methods of compressing the developed face descriptors to make the biometric validation system more suitable for mobile applications. As an additional contribution, we introduce several print-capture datasets that may be used for training and evaluating similar systems for mobile identification and travel documents validation.

Human-robot-interaction using cloud-based speech recognition systems

Progress in natural speech processing has enabled significantly more powerful speech processing systems, primarily due to the use of machine learning technologies. In order to integrate cloud-based speech recognition systems for human-robot interaction, an interface for the voice control of a lightweight robot was developed. The main contribution of this work is the design and implementation of a software interface to recognize commands via cloud-based speech processing and the subsequently conversion into machine-readable code. Requirements for the evaluation of different cloud-services for the control of robots are determined. Furthermore, the control architecture for the robot is modeled and implemented. An example application, which enables users to control robot movements via speech, is realized as a proof of concept and for additional studies. This application includes the basic features of cloud-based speech processing: intent recognition from utterances, slot filling and dialogue-based interaction. Lastly, the influence of background noise on process safety was examined within an experiment. It turns out that a feasible process reliability can be achieved with the system despite the presence of background noises.

ReactionCode: format for reaction searching, analysis, classification, transform, and encoding/decoding

In the past two decades a lot of different formats for molecules and reactions have been created. These formats were mostly developed for the purposes of identifiers, representation, classification, analysis and data exchange. A lot of efforts have been made on molecule formats but only few for reactions where the endeavors have been made mostly by companies leading to proprietary formats. Here, we present ReactionCode: a new open-source format that allows one to encode and decode a reaction into multi-layer machine readable code, which aggregates reactants and products into a condensed graph of reaction (CGR). This format is flexible and can be used in a context of reaction similarity searching and classification. It is also designed for database organization, machine learning applications and as a new transform reaction language.

Programming Diffusion and Localization of DNA Signals in 3D-Printed DNA-Functionalized Hydrogels.

Additive manufacturing enables the generation of 3D structures with predefined shapes from a wide range of printable materials. However, most of the materials employed so far are static and do not provide any intrinsic programmability or pattern-forming capability. Here, a low-cost 3D bioprinting approach is developed, which is based on a commercially available extrusion printer that utilizes a DNA-functionalized bioink, which allows to combine concepts developed in dynamic DNA nanotechnology with additive patterning techniques. Hybridization between diffusing DNA signal strands and immobilized anchor strands can be used to tune diffusion properties of the signals, or to localize DNA strands within the gel in a sequence-programmable manner. Furthermore, strand displacement mechanisms can be used to direct simple pattern formation processes and to control the availability of DNA sequences at specific locations. To support printing of DNA-functionalized gel voxels at arbitrary positions, an open source python script that generates machine-readable code (GCODE) from simple vector graphics input is developed.

Fact – Centered ETL: A Proposal for Speeding Business Analytics up

Abstract In real-time Business Analytics scenarios, like Business Activity Monitoring (BAM) or Operational Intelligence, speeding up ETL is fundamental to provide business users with up-to-date data in order to support decision-making process and to optimize business operations. Conventional ETL processes extract data from heterogeneous operational data sources, transform them according to some predefined semantic and syntactic rules, and finally load results into a new (relational or data warehouse) model, in order to be processed both for business monitoring or analysis. Usually the whole loading process is presented as a technical aspect, far from business facts. In real-time Business Analytics context the traditional ETL approach can still be valid, provided that ETL process has to be designed in a way that circumscribes individual business events and facts and makes them independent one from another. In this context we propose a control-flow- based approach to ETL process modeling, which starts from business facts identification, and represents ETL processes using BPMN notation, which is the foundation for machine-readable code. Our main contribution consists in a proposal for structuring ETL processes and related objects and in its application to a business case.

Monolith: Lessons Learned on the Way to the Market

In 2006 the Imaging and Media Lab at the University of Basel started a first project called Peviar (Permanent Visual Archive) [1, 2]. The aim was to find a digital archival storage solution without the need of periodic migration. The basic approach was to store binary data as 2D barcode on micrographic film. The theoretical, promising results found in Peviar have been further developed. Two subsequent applied science projects concluded in a commercial product called Monolith, which was introduced on the market in 2008.Monolith is a workflow for migration-less preservation of digital data on optical media. It combines the permanence and visual nature of photographic material and the strength of digital imaging technology. The binary information is stored as bit pattern in 2D barcode either on one or multiple film layers. The machine readable code is enriched by human readable metadata in order to describe the archived objects. It also provides detailed descriptions of how to recover the original files. Because of the image based approach, the recovery is not affected by change of technology; digital cameras certainly will be available in the future. Moreover, they will become better and cheaper.

Watchpoint: Bandwagon - Can biometrics-based security stay ahead of the hackers?

Turning our bodily identifiers into machine-readable code makes biometrics-based security systems just as vulnerable to being hacked as anything else, argues the author.

Usable privacy and security for personal information management

The goal is a policy workbench enabling users to create and transform natural language policies into machine-readable code for enforcement and compliance audits.

Practical guide to bar coding for patient medication safety.

Bar coding for the medication administration step of the drug-use process is discussed. FDA will propose a rule in 2003 that would require bar-code labels on all human drugs and biologicals. Even with an FDA mandate, manufacturer procrastination and possible shifts in product availability are likely to slow progress. Such delays should not preclude health systems from adopting bar-code-enabled point-of-care (BPOC) systems to achieve gains in patient safety. Bar-code technology is a replacement for traditional keyboard data entry. The elements of bar coding are content, which determines the meaning; data format, which refers to the embedded data and symbology, which describes the "font" in which the machine-readable code is written. For a BPOC system to deliver an acceptable level of patient protection, the hospital must first establish reliable processes for a patient identification band, caregiver badge, and medication bar coding. Medications can have either drug-specific or patient-specific bar codes. Both varieties result in the desired code that supports patient's five rights of drug administration. When medications are not available from the manufacturer in immediate-container bar-coded packaging, other means of applying the bar code must be devised, including the use of repackaging equipment, overwrapping, manual bar coding, and outsourcing. Virtually all medications should be bar coded, the bar code on the label should be easily readable, and appropriate policies, procedures, and checks should be in place. Bar coding has the potential to be not only cost-effective but to produce a return on investment. By bar coding patient identification tags, caregiver badges, and immediate-container medications, health systems can substantially increase patient safety during medication administration.

Low‐density diffractive optical memories for document security

A low-density diffractive optical write-once, read multiple times memory (WORM) is demonstrated that authenticates and secures documents. Furthermore, by encoding the machine-readable optical code, information can be linked to the document. The proposed solution leads to a reliable, high security machine-readable information carrier that is fully compatible with the requirements of typical security docu- ments, such as payment substrates, fiduciary papers, passports, and identity cards. The information carrier is a juxtaposition of diffractive structures. Data are written permanently in an optical memory; the dif- fractive structure is changed irreversibly through the interaction of a dif- fractive surface with a beam of laser light. A prototype low-density optical memory based on diffractive structures that are resilient against copying or reorigination is realized for use in securing official travel documents. © 1996 Society of Photo-Optical Instrumentation Engineers.

Automated Scanning and Digitizing of Roentgenographs for Documentation and Research

A comprehensive system has been developed for analyzing and reporting the results of total hip arthroplasty. The personal-computer-based system links patient demographic data with digital storage, retrieval, and analysis of roentgenographs. The system consists of a roentgenograph scanner for converting sheet film to digital data, an optical mark reader for patient data input, an archiving system with optical storage, and a physician display station for preoperative planning and postoperative evaluation. Once a roentgenograph has been digitized and stored, the image can be retrieved and manipulated in a manner not possible with the original sheet film. A selected roentgenograph can be brought to full or enlarged scale, enhanced, and overlaid with templates for preoperative planning or for postoperative measurement of changes. In addition, an intelligent database system has been developed for linking patient demographic information with the roentgenographic data. The database system employs uniform criteria and terminology and allows the retrospective study and statistical analysis of comparable cases. Three machine-readable code sheets are used: Form A, Replacement of the Hip; Form B, Hip Prosthesis Reoperation; and Form C, Follow-up. Forms A and B contain information concerning anamnesis, diagnosis, treatment, postoperative course, recovery, and discharge of the patient from the hospital. Form C provides information on physical examination, pain, mobility of the hip, walking ability, and evaluation of the results by the surgeon as well as the patient.

Ink jet printing — Post Office project

The British Post Office uses a phosphorescent machine readable code to enable letters to be sorted automatically. The phosphorescent material is printed in the form of a dot pattern in two rows across the face of the envelope. When activated by passing the latter under an Ultra‐violet light the phospor dots will start to glow and will continue glowing for some time after leaving the activating light source. This afterglow is read by a photodetector and the code pattern is interpreted by the machine electronics which instructs the machine mechanics to place the letter in the correct sortation box.

Storage and Retrieval of SNOP-coded Pathologic Diagnoses Using Offsite Computing and Optical Character Recognizing Systems

A computerized cross-reference system for retrieving autopsy and surgical pathology cases on the basis of case number or diagnosis has been implemented. The system achieves economy and flexibility by using offsite computer service bureaus for job production, eliminating the need for expensive onsite equipment. Coded diagnoses may be typed using the OCR (Optical Character Recognition) font simultaneously with or separately from the clinical documentation. The flexibility of new OCR equipment permits production of machine-readable code sheets with an ordinary pencil and completely eliminates the need for typing. The system produces year-to-date books that list all diagnoses, on an accumulating basis, in alphabetic order by SNOP* topology, morphology, etiology and function, and will be compatible with SNOMed. Because all data are stored on magnetic tape, they may be manipulated and retrieved as desired through user programming. The initial setup cost was dollar 1,000 for programming and testing, and production runs and all report printing cost about dollar 1,000 per year (autopsies and surgical pathology cases), which is about 1.1 cents per diagnosis.

Automated Retrieval of Autopsy Diagnose by Computer Technique

The single most important objection to conventional systems of retrieval of pathological diagnoses is that the coding must be done by hand for each record. The realization that diagnoses are regularly expressed in terms of pathologic process, anatomic site, and modifier of the process, coupled with the extraordinary manipulative capacity of the electronic computer, provides a way to eliminate hand coding of diagnoses. A machine-readable code for all diagnostic words in several autopsy reports has been formulated. By automatic translation of English words to code, and by automatic transfer of code to tape, it is possible to store and to retrieve diagnoses promptly and accurately by machine alone. It is necessary that questions be expressed in, “machine language.” Retrieval can be accomplished either on a generic or a specific basis, so that broad surveys can be made, or a single autopsy containing very specific information can be searched for.