PhagoStat a scalable and interpretable end to end framework for efficient quantification of cell phagocytosis in neurodegenerative disease studies

Quantifying the phagocytosis of dynamic, unstained cells is essential for evaluating neurodegenerative diseases. However, measuring rapid cell interactions and distinguishing cells from background make this task very challenging when processing time-lapse phase-contrast video microscopy. In this study, we introduce an end-to-end, scalable, and versatile real-time framework for quantifying and analyzing phagocytic activity. Our proposed pipeline is able to process large data-sets and includes a data quality verification module to counteract potential perturbations such as microscope movements and frame blurring. We also propose an explainable cell segmentation module to improve the interpretability of deep learning methods compared to black-box algorithms. This includes two interpretable deep learning capabilities: visual explanation and model simplification. We demonstrate that interpretability in deep learning is not the opposite of high performance, by additionally providing essential deep learning algorithm optimization insights and solutions. Besides, incorporating interpretable modules results in an efficient architecture design and optimized execution time. We apply this pipeline to quantify and analyze microglial cell phagocytosis in frontotemporal dementia (FTD) and obtain statistically reliable results showing that FTD mutant cells are larger and more aggressive than control cells. The method has been tested and validated on several public benchmarks by generating state-of-the art performances. To stimulate translational approaches and future studies, we release an open-source end-to-end pipeline and a unique microglial cells phagocytosis dataset for immune system characterization in neurodegenerative diseases research. This pipeline and the associated dataset will consistently crystallize future advances in this field, promoting the development of efficient and effective interpretable algorithms dedicated to the critical domain of neurodegenerative diseases' characterization. https://github.com/ounissimehdi/PhagoStat .

Bladder perforation in a peritoneal dialysis patient

The dysfunction of the catheter in peritoneal dialysis (PD) is a frequent complication. However, perforation of organs are rare, particularly that of the urinary bladder. This requires an early diagnosis and prompt treatment of patients. We report here the case of a 38-year-old woman having end-stage renal disease due to autosomal-dominant polycystic kidney disease treated by PD since November 2000. Three years later, she was treated for Staphylococcal peritonitis. Four months later, she presented with a severe urge to urinate at the time of the fluid exchanges. The biochemical analysis of the fluid from the bladder showed that it was dialysis fluid. Injection of contrast through the catheter demonstrated the presence of a fistula between the bladder and the peritoneal cavity. She underwent cystoscopic closure of the fistulous tract and the PD catheter was removed. Subsequently, the patient was treated by hemodialysis. One month later, a second catheter was implanted surgically after confirming the closure of the fistula. Ten days later, she presented with pain at the catheter site and along the tunnel, which was found to be swollen along its track. The injection of contrast produced swelling of the subcutaneous tunnel but without extravasation of the dye. PD was withdrawn and the patient was put back on hemodialysis. Bladder fistula is a rare complication in PD and diagnosis should be suspected when patient complains of an urge to pass urine during the exchanges, which can be confirmed by contrast study showing presence of dye in the bladder. PD may be possible after the closure of the fistula, but recurrence may occur.

Rehospitalization after kidney transplantation during the first year: length, causes and relationship with long-term patient and graft survival

INTRODUCTION

There is a wide interest in epidemiologic studies assessing different causes of post-kidney transplantation rehospitalization. However, there is a paucity of knowledge on the long-term survival and graft function of rehospitalized kidney transplant recipients during the first year. Knowledge of posttransplant rehospitalization causes may help guide the preventive program at the first year. In our study, we assess causes for hospitalization and investigate the long-term patient and graft survival after non-fatal rehospitalization in kidney recipients during the first year.

MATERIALS AND METHODS

We retrospectively studied the medical histories of 419 kidney transplant recipients whose operations were performed between 1986 and 2009 at Charles Nicolle Hospital, in Tunis, Tunisia. Among these patients, a total of 296 posttransplant rehospitalizations of kidney transplant recipients during the first year occurring in 191 (45.5%) patients were assessed. Clinical characteristics of the patients, including gender, age, reason for kidney failure, weight, height, blood group, length of pretransplant dialysis, immunosuppressive regimen, postoperative complications, the length of hospital stay, transplantation-admission interval, causes of rehospitalizations, graft loss, and mortality rate were reviewed. For donors, these demographics included age, gender, blood group, type of donor (deceased or living), and relationship to the recipient. Because rehospitalizations are possible for more than one cause, the sum of frequencies of rehospitalization causes is more than 100%.

RESULTS

There was 1 rehospitalization in 121 patients, 2 rehospitalizations in 47 patients, 3 rehospitalizations in 15 patients, 4 rehospitalizations in 5 patients, 5 rehospitalizations in 2 patients and 6 rehospitalizations in 1 patient. Rehospitalization was more frequent for diabetic patients without significant association. The causes of rehospitalization were infection in 221 cases (55.5%), renal dysfunction in 106 cases (26%), cardiovascular event in 10 cases (2.4%), and diabetic ketoacidosis in 11 cases (2.7%). The length of hospital stay was 22.5 ± 29.6 days, 20.15 ± 22.16 days, 25 ± 30 days and 23.4 ± 27.5 days, respectively, in the first, second, third, and fifth rehospitalizations. Median hospital stay for all rehospitalizations was between 14 and 16 days. The risk factors of rehospitalization were: use of mycophenolate mofetile (P = .0072), use of cyclosporine (P = .0073), and cytomegalovirus infection (P < .001). There was no significant correlation between rehospitalization and either lost of graft and death.

CONCLUSIONS

During the first year after kidney transplantation, rehospitalization was especially required because of infections and renal dysfunction. The risk factors of rehospitalization were cadaveric graft, use of mycophenolate mofetil, use of cyclosporine, and cytomegalovirus infection. To prevent and minimize rehospitalizations during the first year, a specific preventive program based on infection prevention and graft function monitoring should be established.

Short- and long-term outcomes of living donors in Tunisia: a retrospective study

Despite initiatives to increase cadaveric donation, there is still a shortfall in donor organs. Kidneys from living donors now makes a significant contribution to increasing the number of organs available for transplantation in Tunisia. We performed a retrospective study of 405 kidney transplantations, including 321 (79.3%) from living donors performed from June 1986 to December 2007. We obtained information on only 162 (50.4%), namely, 64 men (39.5%) and 98 women (60.5%), whose mean age at the time of donation was 42.3 ± 12.2 years. Twelve (8.22%) perioperative complications occurred: wound infections (n = 4), pneumothorax (n = 4), phlebitis (n = 1), hematomas (n = 2), and urinary infection (n = 1). The mean follow-up period was 117.4 ± 74.4 months. Hypertension occurred in 42 donors (25.9%) with mean values of 134 ± 20 for systolic and 79 ± 10 for diastolic blood pressure. Twelve donors (7.4%) developed proteinuria (mean proteinuria, 0.08 ± 1.25 g/d). Renal insufficiency was found in 28 donors (19.44%), 2 of whom developed chronic renal failure requiring dialysis at intervals of 36 and 84 months. In both cases, we diagnosed a familial form of focal segmental glomerulosclerosis. Two donors (1.2%) died within 10 years after kidney donation due to senility. The relatively favorable outcomes suggest that living-donor kidney transplantation is an acceptable approach, in view of the superior results it yields in recipients. However, efforts to increase the number of cadaveric donors in Tunisia should be made. It is also important to develop a registry of long-term kidney function after kidney donation.

Dissecting aortic aneurysm in maintenance hemodialysis patients

The dissecting aortic aneurysm (DAA) is a rare pathology that may result in fatal outcome. We report follow up of three cases of DAA patients undergoing maintenance hemo-dialysis who were managed conservatively.

[Peritoneal dialysis in the elderly]

BACKGROUND

Peritoneal dialysis is used more and more as treatment of substitution of the end stage renal disease at the old subjects of more than 65 years and its effectiveness and its good tolerance were shown.

AIMS

To study the epidemiological and clinical profile of these patients, the indication and the advantages of PD, the various complications, the elements of prognostic and the survival of the patients and the technique.

METHODS

We made a retrospective study including a series of 13 old patients of more than 65 years and treaties by PD during the period extending from the 1983/2/11 to the 2005/12/31. They are 10 men and 3 women on average of 70 +/- 3.1 year and representing 3.62% of the totality of the patients. The diabetic and vascular nephropathies represent the first cause of ESRD.

RESULTS

The PD was used of first intention at 53.84% of the patients whose majority suffers from a malnutrition due to advanced age, anorexia, psychological disorders, bad dental state and uraemia. The cardiovascular complications are frequent in this age bracket explaining heavy morbidity. The pulmonary and urinary infectious complications are also frequent. The more frequent peritonitis compared to the literature, are comparable between the 2 age brackets < and = with 65 years and the mode of PD (APD or CAPD). The time separating the beginning from PD and which has occurred of the peritonitis is shorter in APD. The lesions of renal osteodystrophy are found among 6 patients: 3 cases of hyperparathyroidism and 3 cases of adynamic osteopathy. The return in HD is rare due to dysfunction of the catheter. The survival of the patients is 92.8% at 1 year and 60.8% at 5 years; that of the technique is worse with 88.3% at 1 year and 33.7% at 5 years Eight patients died (61.5%) because of cardiovascular diseases and of the infections. We found a correlation statistically significant between the survival of the patients and the mode of PD, on the other hand any correlation was not found with the age or the sex. The survival of the technique is not correlated to a significant degree with the age, the sex and the mode of PD.

CONCLUSION

Peritoneal dialysis is used of first intention at more half of the old subjects and remains a last recourse for haemodialysis which have an initially vascular problem. The family support and the good nurse allow these patients to adhere well to the technique. The cardiovascular diseases and the infections are responsible for heavy morbi-mortality. The survival of the technique is worse than that of the patients.

Kidney transplantation: Charles Nicolle Hospital experience

The aim of our retrospective study was to analyze the short- and long-term follow-up of 298 renal transplantations performed between June 1986 and May 2005. All were first transplantations except 4 cases, with 54 from cadaveric and 244 from living donors. The recipients included 196 males and 102 females of overall mean age of 31.21 +/- 8.9 years (range, 16-61 years). A combination of prednisolone and azathioprine was presented for 212 patients or mycophenolate mofetil for 86 patients. Polyclonal or monoclonal antibodies were used as induction therapy in 183 cases. Cyclosporine was administered to 188 cases and tacrolimus only to 16. HLA matching was 0 mismatches (MM) in 65 cases; 1 or 2 MM in 113; 3 MM in 99; and > or =4 MM in 21. Acute tubular necrosis occurred in 45 cases. One hundred eighteen patients experienced at least 1 acute rejection episode: 102 cases (41.8%) among living and 16 (29.6%) among cadaveric kidneys donor (P = .0007). The actuarial patient and graft survival rates at 1, 5, 10, 15, and 20 years were 95.9%, 87.4%, 77.5%, 65.6%, and 60.8%, and 94.9%, 84.5%, 75.4%, 65.4%, and 53%, respectively. Sixty-three patients died and 72 patients returned to dialysis. Our results were comparable to experienced centers. However, the member of kidney transplantations does not match the increased number of patients on renal replacement therapy. It is advisable to promote obtaining organs from brain-dead donors.

Impact of initial blood pressure on the mortality of diabetics undergoing renal replacement therapy

The aim of this study was to determine the blood pressure (BP) profiles and their impact on mortality among a cohort of uremic diabetics treated by hemodialysis. The studied population includes all type II diabetics starting hemodialysis for end-stage renal disease between 1990 and 1996. There were 221 patients (144 men, 77 women) aged from 37 to 78 years, were all followed until death or December 2003 without any censored data. Survival analysis to identify predictors of death was performed using the actuarial method, Cox proportional model, including systolic, diastolic, mean, and pulse blood pressures (SBP, DBP, MBP, PP). One hundred seventy-eight patients (80.5%) were hypertensive at the start of dialysis. Hypertension preceded the diagnosis of diabetes in eight cases (4.5%); 154 patients (86.5%) received antihypertensive drugs and only 23 (14.9%) had well-controlled hypertension. Our population was subdivided into four groups according to their BP levels at the time of beginning of dialysis; G1 (19.5%): normal BP (SBP [90 to 140] and DBP [60 to 90]); G2 (30.3%): Hypertension stage 1 (SBP [140 to 160] and/or DBP [90 to 100]); G3 (32.1%): hypertension stage 2 (SBP [160 to 180] and/or DBP [100 to 110]); G4 (18.1%) hypertension stage 3 (SBP [180 to 220] and/or DBP [110 to 120]). Mean age and comorbidities were similar among the four groups. During a cumulative follow-up period of 872 patient-years, 191 patients died, representing a rate of 21.9 per 100 patient-years; 20.42% of these deaths occurred during the first 3 months of dialysis. Normotensive patients showed lower survival rates without any significant difference in comparison with those of other hypertensive groups. None of the initial BP parameters (SBP, DBP, PP, MBP, hypertension stages) seemed to influence early or global mortalities, which were rather related to the urgent onset of renal replacement therapy, to age, to serum albumin, and to the score of associated morbidities. We conclude that mortality of our hemodialyzed diabetics was not influenced by the blood pressure parameters recorded at the onset of dialysis.