Minimal data requirement for realistic endoscopic image generation with Stable Diffusion

PURPOSE

Computer-assisted surgical systems provide support information to the surgeon, which can improve the execution and overall outcome of the procedure. These systems are based on deep learning models that are trained on complex and challenging-to-annotate data. Generating synthetic data can overcome these limitations, but it is necessary to reduce the domain gap between real and synthetic data.

METHODS

We propose a method for image-to-image translation based on a Stable Diffusion model, which generates realistic images starting from synthetic data. Compared to previous works, the proposed method is better suited for clinical application as it requires a much smaller amount of input data and allows finer control over the generation of details by introducing different variants of supporting control networks.

RESULTS

The proposed method is applied in the context of laparoscopic cholecystectomy, using synthetic and real data from public datasets. It achieves a mean Intersection over Union of 69.76%, significantly improving the baseline results (69.76 vs. 42.21%).

CONCLUSIONS

The proposed method for translating synthetic images into images with realistic characteristics will enable the training of deep learning methods that can generalize optimally to real-world contexts, thereby improving computer-assisted intervention guidance systems.

Removal of humic substances from aqueous solutions by the coagulation process

Laboratory studies were carried out on model solutions prepared on a tap-water base. The content of humic substances was 20 mg l(-1). Hardness and pH parameters of these solutions were altered using appropriate reagents. The coagulation process was run in a conventional manner using the two most popular coagulants, i.e. aluminium sulphate and iron chloride III. Optimum doses of coagulants, as well as the effect of hardness and pH of the purified solutions on coagulation process effectiveness, were determined. The efficacy of coagulation using the optimum doses of coagulants was the greatest for solutions having the highest hardness. The coagulation process progressed most efficiently in the pH range from 5 to 6. In order to describe a coagulation process with a mathematical model, the concentration of humic substances was changed from 5 to 30 mg l(-1). The results of these experiments served as a basis for the determination of the dependence of coagulant dose on the initial concentration of solutions.

Therapeutic effects of cysteine protease inhibition in allergic lung inflammation: inhibition of allergen-specific T lymphocyte migration

OBJECTIVE AND DESIGN

We have evaluated the effects of the broad-spectrum cysteine protease inhibitor E64 on allergic lung inflammation in the mouse ovalbumin model of human asthma. We have also characterised membrane-associated cathepsin enzyme activity on a range of cell types.

MATERIALS

Balb/C mice, E64 and CA074, various cell lines.

TREATMENT

E64 was administered by subcutaneous minipump into ovalbumin-sensitised mice prior to intranasal ovalbumin challenge. The effect of E64 on ovalbumin-induced inflammation in vivo and ovalbumin-specific T cell proliferation in vitro and ex vivo was examined. Membrane-associated cathepsin activity on various cell types was measured.

RESULTS

E64 treatment (0.36-0.48 mg/day) led to a significant reduction in eosinophil numbers and lung weights in the mouse model. Histological examination of lungs confirmed the anti-inflammatory effect. E64 greatly reduced ovalbumin-specific T cell numbers in the lymph nodes draining the lung following intranasal challenge whilst an accumulation of these T cells was found in the 'priming' lymph nodes. An analysis of various cells involved in lymphocyte priming and migration revealed that monocytes, dendritic cells and endothelial cells express high levels of membrane-associated cathepsin B activity.

CONCLUSIONS

Since E64 is not cell permeable and does not inhibit antigen-induced T cell proliferation in vitro or in vivo, the data indicate that membrane-associated cysteine proteases, possibly cathepsin B, may regulate T lymphocyte migration in vivo.

Chromosomal abnormalities in patients with Hodgkin's disease: evidence for frequent involvement of the 14q chromosomal region but infrequent bcl-2 gene rearrangement in Reed-Sternberg cells

BACKGROUND

Rearrangements of the bcl-2 gene (also known as BCL2) have been detected in up to 40% of cases of Hodgkin's disease, and it has been speculated that such rearrangements may have a role in the pathogenesis of Hodgkin's disease.

PURPOSE

The purposes of this study were (a) to assess the frequency of clonal chromosomal abnormalities in Hodgkin's disease, (b) to identify recurrent changes, (c) to determine whether the bcl-2 gene rearrangement was present in Reed-Sternberg cells (the neoplastic cells of Hodgkin's disease) and their variants, and (d) to analyze whether the presence of t(14;18) translocations in Reed-Sternberg cells explains the observed bcl-2 gene rearrangements in Hodgkin's disease.

METHODS

A cytogenetic study was performed on biopsy specimens from 28 consecutive untreated patients with Hodgkin's disease. The same patients were analyzed for bcl-2 gene rearrangement by a polymerase chain reaction (PCR) technique. To ascertain whether the abnormal karyotypes were present in and restricted to Reed-Sternberg cells, we also performed in situ hybridization with chromosome-specific probes.

RESULTS

Abnormal metaphases were identified in 23 of the 28 patients. In 11 patients, the chromosome 14q region was abnormal; in six of these patients, there was involvement of the 14q32 region that comprises the gene encoding for heavy-chain immunoglobulin. Only one patient had a t(14;18) translocation, whereas almost 40% of these 28 patients showed bcl-2 gene rearrangements by a PCR method. The in situ hybridization method showed that the abnormal karyotype was present in and restricted to Reed-Sternberg cells.

CONCLUSIONS

We conclude that the majority of cases of Hodgkin's disease contain a clonal population with an abnormal karyotype, comprising the Reed-Sternberg cells. The q32 region of chromosome 14 is frequently involved, but a t(14;18) translocation is extremely infrequent. The occurrence of a bcl-2 gene rearrangement in Hodgkin's disease most likely results from the presence of sporadic, small bystander B lymphocytes that carry the translocation and that also can be frequently detected in reactive lymphoid tissue such as tonsils. Also, a range of different chromosomal translocations may provide growth or survival advantages to Reed-Sternberg cells.

Biology of Hodgkin's disease

The biology of Hodgkin's disease is one of the most intriguing subjects in lymphoma research. The presence of only a small proportion of neoplastic cells and a vast majority of reactive cells reflects the presence of complex interactions between these cell types. In this paper we discuss findings indicating that Reed-Sternberg cells may be virally transformed cells that have evaded a cytotoxic immune response and induce an ineffective delayed type hypersensitivity reaction.