Accuracy of Information given by ChatGPT for patients with Inflammatory Bowel Disease in relation to ECCO Guidelines

INTRODUCTION

As acceptance of AI platforms increases, more patients will consider these tools as sources of information. The ChatGPT architecture utilizes a neural network to process natural language, thus generating responses based on the context of input text. The accuracy and completeness of ChatGPT3.5 in the context of Inflammatory Bowel Disease remains unclear.

METHODS

In this prospective study, 38 questions worded by IBD patients were inputted into ChatGPT3.5. The following topics were covered: 1) CD, UC and malignancy, 2) maternal medicine 3) infection and vaccination 4) complementary medicine. Responses given by Chat GPT were assessed for accuracy (1 - completely incorrect to 5 - completely correct) and completeness (3-point Likert scale; range 1 - incomplete to 3 - complete) by 14 expert gastroenterologists, in comparison with relevant ECCO guidelines.

RESULTS

In terms of accuracy, most replies (84.2%) had a median score of ≥4 (IQR:2) and a mean score of 3.87 (SD: +/- 0.6). For completeness, 34.2% of the replies had a median score of 3 and 55.3 % had a median score of between 2 and <3. Overall, the mean rating was 2.24 (SD: +/- 0.4, Median:2 IQR :1). Though group 3 and 4 had a higher mean for both accuracy and completeness, there was no significant scoring variation between the 4 question groups (Kruskal-Wallis test p:>0.05). However, statistical analysis for the different individual questions revealed a significant difference both for accuracy (p<0.001) and completeness (p<0.001). The questions which rated the highest for both accuracy and completeness were related to smoking, while the lowest rating was related to screening for malignancy and vaccinations especially in the context of immunosuppression and family planning.

CONCLUSION

This is the first study to demonstrate the capability of an AI-based system to provide accurate and comprehensive answers to real-world patient queries in IBD. AI systems may serve as a useful adjunct for patients, in addition to standard of care in clinic and validated patient information resources. However, responses in specialist areas may deviate from evidence-based guidance and the replies need to give more firm advice.

Non-fatal case of Crimean-Congo haemorrhagic fever imported into the United Kingdom (ex Bulgaria), June 2014

Crimean-Congo haemorrhagic fever (CCHF) was diagnosed in a United Kingdom traveller who returned from Bulgaria in June 2014. The patient developed a moderately severe disease including fever, headaches and petechial rash. CCHF was diagnosed following identification of CCHF virus (CCHFV) RNA in a serum sample taken five days after symptom onset. Sequence analysis of the CCHFV genome showed that the virus clusters within the Europe 1 clade, which includes viruses from eastern Europe.

Analysis of errors in finished DNA sequences: the surfactin operon of Bacillus subtilis as an example

Increased productivity in DNA sequencing would not be valid without a straightforward detection and estimation of errors in finished sequences. The sequence of the surfactin operon from Bacillus subtilis was obtained by two different groups and by chance we were also working on the same chromosome region. Taking advantage of this situation we report in this paper, the number and nature of errors found in the overlapping part of the DNA sequences obtained by the three laboratories. The coincidence of some of the errors with compression in sequence ladders and with secondary DNA structures as well as the detection of frameshift errors using computer programs, are demonstrated. Finally we discuss the definition of a new sequencing strategy that might minimize both the error rate and the cost of sequencing.

Purification and properties of a periplasmic aminoendopeptidase from Escherichia coli

A periplasmic aminoendopeptidase from Escherichia coli has been purified to hemogeneity. It is a monomer of molecular weight 45000 and containing one -- SH group that is necessary for catalytic activity. The study of its substrate specificity indicated that the enzyme has both aminopeptidase and endopeptidase activity. The pH optimum for L-alanine p-nitroanilide hydrolysis is between 7 and 7.5 and that for 125I-labeled casein proteolysis between 7.3 and 7.6. The activation energy for the hydrolysis of L-anine p-nitroanilide was calculated to be 5.3 kcal X mol-1 (22.2 kJ X mol-1).