Patient information extraction in digitized radiography

Digital imagery is gradually replacing the traditional radiograph with the development of digital radiography and film scanner. This report presents a new method to extract the patient information number (PIN) field automatically from the film-scanned image using image analysis technique. To evaluate the PIN field extraction algorithm, 2 formats of label acquired from 2 different hospitals are tested. Given the available films with no constraints on the way the labels are written and positioned, the correct extraction rates are 73% and 84%, respectively. This extracted PIN information can link with Radiology Information System (RIS) or Hospital Information System (HIS), and the image scanned from the film then can be filed into the database automatically. The efficiency this method offers can simplify greatly the image filing process and improve the user friendliness of the overall image digitization system. Moreover, compared with the bar code reader, it solves the automatic information input problem in a very economical way. The authors believe the success of this technique will benefit the development of the PACS (Picture Archiving and Communication System) and teleradiology.

Accuracy of on-line databases in determining vital status

Ascertainment of vital status is important for epidemiological and clinical trial research. Two free databases based on the Social Security Administration Death Master File have become available on the Internet. The accuracy of these databases is unknown. A cohort of 124 patients known to be dead and a cohort of 203 patients not known to be dead were identified. The on-line databases were searched with both of these cohorts following a specific search algorithm. The results for both on-line databases were identical. The optimal algorithm had a sensitivity of 0.82 (95% confidence interval 0.74-0.89) for identification of deaths. The sensitivity excluding deaths that occurred during the first year of life (n = 118) was 0.86 (95% CI 0.79-0.92). The specificity was 1.00. This study found that free and convenient on-line databases based on the Social Security Administration Death Master File can be useful in the accurate ascertainment of vital status.

[The approaches to factors which cause medication error--from the analyses of many near-miss cases related to intravenous medication which nurses experienced]

Given the complexity of the intravenous medication process, systematic thinking is essential to reduce medication errors. Two thousand eight hundred cases of 'Hiyari-Hatto' were analyzed. Eight important factors which cause intravenous medication error were clarified as a result. In the following I summarize the systematic approach for each factor. 1. Failed communication of information: illegible handwritten orders, and inaccurate verbal orders and copying cause medication error. Rules must be established to prevent miscommunication. 2. Error-prone design of the hardware: Look-alike packaging and labeling of drugs and the poor design of infusion pumps cause errors. The human-hardware interface should be improved by error-resistant design by manufacturers. 3. Patient names similar to simultaneously operating surgical procedures and interventions: This factor causes patient misidentification. Automated identification devices should be introduced into health care settings. 4. Interruption in the middle of tasks: The efficient assignment of medical work and business work should be made. 5. Inaccurate mixing procedure and insufficient mixing space: Mixing procedures must be standardized and the layout of the working space must be examined. 6. Time pressure: Mismatch between workload and manpower should be improved by reconsidering the work to be done. 7. Lack of information about high alert medications: The pharmacist should play a greater role in the medication process overall. 8. Poor knowledge and skill of recent graduates: Training methods and tools to prevent medication errors must be developed.

On the use of EEG features towards person identification via neural networks

Person identification based on spectral information extracted from the EEG is addressed in this work a problem that has not yet been seen in a signal processing framework. Spectral features are extracted non-parametrically from real EEG data recorded from healthy individuals. Neural network classification is applied on these features using a Learning Vector Quantizer in an attempt to experimentally investigate the connection between a person's EEG and genetically specific information. The proposed method, compared with previously proposed methods, has yielded encouraging correct classification scores in the range of 80% to 100% (case-dependent). These results are in agreement with previous research showing evidence that the EEG carries genetic information.

[Solitary mastocytoma associated with newborn infants' identification bracelets]

We report the case of solitary mastocytoma, which appeared under the identification bracelet on the left wrist of a 2-day-old newborn infant. Stroking the lesion elicited Darier's sign. Histopathologic examination of a punch biopsy specimen revealed a dense mast cell infiltrate of the dermis, which was confirmed by Giemsa and toluidine blue stains. We believe that this case represents and unusual reactive process.

Federation of the Person Identification Service between enterprises

The Person Identification Service (PIDS) is a standard that has been adopted by the Object Management Group for managing identities of persons within a particular domain. That standard includes an interface that supports the ability to connect multiple PIDS servers together in a federated manner. The specification leaves great flexibility as to how to accomplish the federation. In this paper, we examine some of the federated approaches being considered by the Government Computer-based Patient Record Framework (G-CPR) project and discuss their advantages and disadvantages and the details of a specific, scalable approach to federation.

[Genetic maternal-child identification with DNA]

BACKGROUND

Classical methods for newborn identification cannot establish a true biological nexus between mother and newborn, and hence they have been widely criticized. Therefore, a pilot study on a mother-infant genetic identification program (PROIGMI) has been started in order to ensure the determination of a biological relationship between mother and newborn in cases of vaginal delivery, caesarean birth or fetal autopsies.

MATERIAL AND METHOD

In the delivery room and after informed consent, a total of 100 blood samples from mother/newborn couples were taken and deposited on clean and sterile paper supports. DNA from a total of 20 mother/newborn couples was studied by PCR techniques, being able to unequivocally establish the biological relationship in all cases, even when using minimal amounts of DNA.

RESULTS

Blood samples collection does not show differences regarding the kind of birth (delivery, cesarean). The protocol used is easy and fast, and does not employ materials not known for health care professionals. Minimal amounts of blood yield enough DNA to obtain conclusive inclusion probabilities.

CONCLUSIONS

The use of DNA allows to stablish the so called biological truth in cases of doubt or where necessary; with the use of medical protocols these studies can be completed in 6 to 8 hours using small amounts of DNA (5 microliters).

Contact dermatitis from polyvinyl chloride identification bands

One case of allergic contact dermatitis caused by polyvinyl chloride identification bands is reported. Patch tests in this case were positive for three of four bands tested, resorcinol monobenzoate, and benzoyl peroxide. Manufacturers could not confirm the presence of resorcinol monobenzoate and benzoyl peroxide in their respective bands. It is concluded that the patient's reaction was allergic and due to an unknown chemical(s) in the bands. Additives may comprise a significant portion of polyvinyl chloride bands. These components may be leachable, and thus lead to contact sensitization with subsequent allergy. Although patch testing is an excellent first step in determining the offending allergen, specialized chemical analysis may be required for exact identification.

How to find who

Master person indexes are considered the cornerstone of many health care systems integration projects. Although a fundamental first step, mastering the master person index is an IT challenge when organizations expand.

Malaysian antenatal risk coding and the outcome of pregnancy

OBJECTIVE

Measure the effectiveness of the colour coding system in Malaysia for the prediction of risk in pregnancy.

METHOD

Cohort study of records and interviews of 253/279 women examined at first antenatal visit.

RESULTS

Nurses' final coding showed poor concordance with guidelines; recoding produced a predictive value of high risk of 48%; 25% of those with low risk had 50% of complications. Complication rates were moderate and intervention rates low. The mothers had little appreciation of risk and preferred home delivery. Home deliveries gave excellent results except for the 17% requiring transfer to hospital during labour or delivery.

CONCLUSION

The coding system is ineffective with Malaysia's relatively low reproductive risk. Women require more personalised counseling about risk to make appropriate choices. Better results depend on simpler but consistent selection for hospital delivery using reproductive history, combined with better communication and transport systems for home deliveries and a reorientation within hospitals to rapid emergency care.

Practical methods to improve transfusion safety by using novel blood unit and patient identification systems

Human error is a leading cause of transfusion-associated death. Many of these events are associated with a failure to comply with established unit-recipient identification protocols. Although several dedicated systems designed to minimize this problem are currently available, none of them have been sufficiently challenged by the various conditions that exist in diverse clinical settings. However, data available for computer-based recognition procedures and for a disposable blood bag combination lock, which precludes access to the blood before it is properly identified, are encouraging.

A chip card for patients with diabetes

DIABCARD provides the specification for the core of a Chip Card Based Medical Information System (CCMIS) for the treatment of patients with chronic diseases. It will provide an instrument for assessing health care services, improve the links between health care providers and set up communication between the different levels of health care. It will therefore improve the quality of care and thus the life of patients with chronic diseases. DIABCARD concentrates on diabetes at the moment, the concept of the diabetes chip card will, however, be extendable to other chronic diseases.