ABSTRACT
We propose an automated, explainable artificial intelligence (xAI) system for age-related macular degeneration (AMD) diagnosis. Mimicking the physician's perceptions, the proposed xAI system is capable of deriving clinically meaningful features from optical coherence tomography (OCT) B-scan images to differentiate between a normal retina, different grades of AMD (early, intermediate, geographic atrophy (GA), inactive wet or active neovascular disease [exudative or wet AMD]), and non-AMD diseases. Particularly, we extract retinal OCT-based clinical imaging markers that are correlated with the progression of AMD, which include: (i) subretinal tissue, sub-retinal pigment epithelial tissue, intraretinal fluid, subretinal fluid, and choroidal hypertransmission detection using a DeepLabV3+ network; (ii) detection of merged retina layers using a novel convolutional neural network model; (iii) drusen detection based on 2D curvature analysis; (iv) estimation of retinal layers' thickness, and first-order and higher-order reflectivity features. Those clinical features are used to grade a retinal OCT in a hierarchical decision tree process. The first step looks for severe disruption of retinal layers' indicative of advanced AMD. These cases are analyzed further to diagnose GA, inactive wet AMD, active wet AMD, and non-AMD diseases. Less severe cases are analyzed using a different pipeline to identify OCT with AMD-specific pathology, which is graded as intermediate-stage or early-stage AMD. The remainder is classified as either being a normal retina or having other non-AMD pathology. The proposed system in the multi-way classification task, evaluated on 1285 OCT images, achieved 90.82% accuracy. These promising results demonstrated the capability to automatically distinguish between normal eyes and all AMD grades in addition to non-AMD diseases.
ABSTRACT
Appropriate treatment of bladder cancer (BC) is widely based on accurate and early BC staging. In this paper, a multiparametric computer-aided diagnostic (MP-CAD) system is developed to differentiate between BC staging, especially T1 and T2 stages, using T2-weighted (T2W) magnetic resonance imaging (MRI) and diffusion-weighted (DW) MRI. Our framework starts with the segmentation of the bladder wall (BW) and localization of the whole BC volume (Vt) and its extent inside the wall (Vw). Our segmentation framework is based on a fully connected convolution neural network (CNN) and utilized an adaptive shape model followed by estimating a set of functional, texture, and morphological features. The functional features are derived from the cumulative distribution function (CDF) of the apparent diffusion coefficient. Texture features are radiomic features estimated from T2W-MRI, and morphological features are used to describe the tumors' geometric. Due to the significant texture difference between the wall and bladder lumen cells, Vt is parcelled into a set of nested equidistance surfaces (i.e., iso-surfaces). Finally, features are estimated for individual iso-surfaces, which are then augmented and used to train and test machine learning (ML) classifier based on neural networks. The system has been evaluated using 42 data sets, and a leave-one-subject-out approach is employed. The overall accuracy, sensitivity, specificity, and area under the receiver operating characteristics (ROC) curve (AUC) are 95.24%, 95.24%, 95.24%, and 0.9864, respectively. The advantage of fusion multiparametric iso-features is highlighted by comparing the diagnostic accuracy of individual MRI modality, which is confirmed by the ROC analysis. Moreover, the accuracy of our pipeline is compared against other statistical ML classifiers (i.e., random forest (RF) and support vector machine (SVM)). Our CAD system is also compared with other techniques (e.g., end-to-end convolution neural networks (i.e., ResNet50).
Subject(s)
Multiparametric Magnetic Resonance Imaging , Urinary Bladder Neoplasms , Diffusion Magnetic Resonance Imaging , Humans , Magnetic Resonance Imaging , ROC Curve , Retrospective Studies , Support Vector Machine , Urinary Bladder Neoplasms/diagnostic imagingABSTRACT
This study proposes a novel computer assisted diagnostic (CAD) system for early diagnosis of diabetic retinopathy (DR) using optical coherence tomography (OCT) B-scans. The CAD system is based on fusing novel OCT markers that describe both the morphology/anatomy and the reflectivity of retinal layers to improve DR diagnosis. This system separates retinal layers automatically using a segmentation approach based on an adaptive appearance and their prior shape information. High-order morphological and novel reflectivity markers are extracted from individual segmented layers. Namely, the morphological markers are layer thickness and tortuosity while the reflectivity markers are the 1st-order reflectivity of the layer in addition to local and global high-order reflectivity based on Markov-Gibbs random field (MGRF) and gray-level co-occurrence matrix (GLCM), respectively. The extracted image-derived markers are represented using cumulative distribution function (CDF) descriptors. The constructed CDFs are then described using their statistical measures, i.e., the 10th through 90th percentiles with a 10% increment. For individual layer classification, each extracted descriptor of a given layer is fed to a support vector machine (SVM) classifier with a linear kernel. The results of the four classifiers are then fused using a backpropagation neural network (BNN) to diagnose each retinal layer. For global subject diagnosis, classification outputs (probabilities) of the twelve layers are fused using another BNN to make the final diagnosis of the B-scan. This system is validated and tested on 130 patients, with two scans for both eyes (i.e. 260 OCT images), with a balanced number of normal and DR subjects using different validation metrics: 2-folds, 4-folds, 10-folds, and leave-one-subject-out (LOSO) cross-validation approaches. The performance of the proposed system was evaluated using sensitivity, specificity, F1-score, and accuracy metrics. The system's performance after the fusion of these different markers showed better performance compared with individual markers and other machine learning fusion methods. Namely, it achieved [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text], respectively, using the LOSO cross-validation technique. The reported results, based on the integration of morphology and reflectivity markers and by using state-of-the-art machine learning classifications, demonstrate the ability of the proposed system to diagnose the DR early.
Subject(s)
Diabetic Retinopathy/diagnostic imaging , Retina/diagnostic imaging , Early Diagnosis , Humans , Image Processing, Computer-Assisted , Machine Learning , Neural Networks, Computer , Tomography, Optical CoherenceABSTRACT
Cardiac magnetic resonance (MR) imaging is one of the most rigorous form of imaging to assess cardiac function in vivo. Strain analysis allows comprehensive assessment of diastolic myocardial function, which is not indicated by measuring systolic functional parameters using with a normal cine imaging module. Due to the small heart size in mice, it is not possible to perform proper tagged imaging to assess strain. Here, we developed a novel deep learning approach for automated quantification of strain from cardiac cine MR images. Our framework starts by an accurate localization of the LV blood pool center-point using a fully convolutional neural network (FCN) architecture. Then, a region of interest (ROI) that contains the LV is extracted from all heart sections. The extracted ROIs are used for the segmentation of the LV cavity and myocardium via a novel FCN architecture. For strain analysis, we developed a Laplace-based approach to track the LV wall points by solving the Laplace equation between the LV contours of each two successive image frames over the cardiac cycle. Following tracking, the strain estimation is performed using the Lagrangian-based approach. This new automated system for strain analysis was validated by comparing the outcome of these analysis with the tagged MR images from the same mice. There were no significant differences between the strain data obtained from our algorithm using cine compared to tagged MR imaging. Furthermore, we demonstrated that our new algorithm can determine the strain differences between normal and diseased hearts.
Subject(s)
Heart Diseases/diagnostic imaging , Heart Ventricles/diagnostic imaging , Heart/diagnostic imaging , Magnetic Resonance Imaging, Cine , Algorithms , Animals , Heart/physiopathology , Heart Diseases/physiopathology , Heart Ventricles/pathology , Humans , Image Interpretation, Computer-Assisted , Image Processing, Computer-Assisted , Mice , Neural Networks, ComputerABSTRACT
AIMS: To report the prevalence of the metabolic syndrome in Arab men and women using the new International Diabetes Federation (IDF) criteria, and to compare this with the prevalence using the 1999 World Health Organization (WHO) and 2001 National Cholesterol Education Program Adult Treatment Panel III (NECP ATPIII) definitions. METHODS: The study involved 863 subjects (343 men and 520 women) aged > or = 40 years living in Tunis, Tunisia, taken from an initially randomized, population sample. RESULTS: The prevalence of the metabolic syndrome using the IDF criteria was found to be 45.5%; 55.8% in women and 30.0% in men (P < 0.001), higher than the rates of 28.7% (WHO) and 24.3% (NECP ATPIII) using the previous definitions. Using all the definitions, the prevalence was higher in women than in men predominantly because of significant differences in central obesity and high-density lipoprotein (HDL) cholesterol and, to a lesser extent, hypertension. CONCLUSION: The increased prevalence using the IDF criteria compared with the 1999 WHO criteria and the 2001 NCEP ATPIII definitions is striking and has huge implications for public health worldwide. The major reason for the higher rate using the new definition seems to be the predominant focus placed on central obesity. Using tighter criteria for fasting glycaemia has also played a factor. The question remains as yet unanswered as to whether the new IDF criteria are better at predicting hard outcomes such as diabetes mellitus and cardiovascular diseases.
Subject(s)
Metabolic Syndrome/epidemiology , Aged , Analysis of Variance , Arabs , Chi-Square Distribution , Female , Humans , Male , Metabolic Syndrome/diagnosis , Metabolic Syndrome/ethnology , Middle Aged , Practice Guidelines as Topic , Prevalence , Tunisia/epidemiology , World Health OrganizationABSTRACT
The therapeutic value of doxorubicin (DOX) as anticancer antibiotic is limited by its cardiotoxicity. The implication of natural phenolic acids in the prevention of many pathologic diseases has been reported. Herein, the ability of p-coumaric (PC) acid, a member of phenolic acids, to protect rat's heart against DOX-induced oxidative stress was investigated. Three main groups of albino rats were used; DOX, PC, and PC plus DOX-receiving animals. Corresponding control animals were also used. DOX was administered i.p. in a single dose of 15mgkg(-1). PC alone, in a dose of 100mgkg(-1), was orally administered for five consecutive days. In PC/DOX group, rats received PC 5 days prior to DOX. DOX-induced high serum levels of lactic dehydrogenase (LDH) and creatine phosphokinase (CPK), were reduced significantly by PC administration, compared to DOX-receiving rats. Pretreatment with PC ameliorated the cardiac content of glutathione (GSH), and superoxide dismutase (SOD) & catalase (CAT) activities, compared to DOX-receiving rats. On the other hand, accumulation of cardiac content of MDA significantly decreased following PC pretreatment, compared to DOX-treated rats. The data presented here indicate that PC protects rats hearts against DOX-induced oxidative stress in the heart. It may be worthy to consider the usefulness of PC as adjuvant therapy in cancer management.
Subject(s)
Anti-Bacterial Agents/pharmacology , Coumaric Acids/pharmacology , Doxorubicin/pharmacology , Heart/drug effects , Myocardium/metabolism , Oxidative Stress/drug effects , Animals , Antioxidants/metabolism , Catalase/metabolism , Creatine Kinase/metabolism , Glutathione/metabolism , L-Lactate Dehydrogenase/metabolism , Lipid Peroxidation/drug effects , Malondialdehyde/metabolism , Oxidation-Reduction , Propionates , Proteins/metabolism , Rats , Superoxide Dismutase/metabolismABSTRACT
OBJECTIVE: To determine the predictive risk factors for the development of type 2 diabetes mellitus (DM) or impaired glucose tolerance (IGT) in a normoglycemic population. RESEARCH DESIGN AND METHODS: This is a ten-year prospective study in a randomly selected urban population including 1835 subjects aged >=30 years living in Tunis, 1460 were normoglycemic according to 2 hours blood glucose WHO criteria, and 701 among them attended the follow-up assessment ten years later. Subjects with impaired glucose tolerance (IGT) were excluded. Subjects underwent a physical examination including weight/height, iliac circumference (IC) and blood pressure measurements. Fasting and 2-hour venous blood sampling, after a 75 g oral glucose load, were performed for the measurement of plasma glucose (G(0), G(2h)), insulin (I(0), I(2h)), total cholesterol (TC) and glycated hemoglobin (HbA(1c)) levels. RESULTS: Out of the 701 normoglycemic subjects in 1985, 77 developed diabetes and 130 impaired glucose tolerance after 10 years, giving a mean annual incidence rate of 1.1% for diabetes and 1.85% for IGT. Univariate analysis showed that risk factors for diabetes were age, BMI, IC, SBP, G(0) and total cholesterol in both sexes, I(0) and I(2h) in men only and DBP G(2h) and HbA(1c) in women only. Risk factors for IGT were BMI, IC and G(2h) in both sexes, I(2h) in men only and G(0) in women only. Multivariate analysis revealed that BMI, G(0) and G(2h) were independent risk factors for conversion to diabetes or IGT in both sexes, but HbA(1c) and IC were risk factors only in men. CONCLUSION: Early screening and prevention of diabetes must focus on obese subjects, especially those with central fat distribution, and those with moderate increase in fasting and/or two-hour blood glucose levels within the normal range.
Subject(s)
Blood Glucose/metabolism , Diabetes Mellitus, Type 2/epidemiology , Glucose Intolerance/epidemiology , Glucose Tolerance Test , Urban Population/statistics & numerical data , Adult , Analysis of Variance , Blood Pressure , Body Mass Index , Female , Follow-Up Studies , Glycated Hemoglobin/analysis , Humans , Incidence , Male , Middle Aged , Multivariate Analysis , Predictive Value of Tests , Prospective Studies , Reference Values , Risk Factors , Time Factors , Tunisia/epidemiologyABSTRACT
The authors report the results of a randomised trial using a converting enzyme inhibitor in 40 microalbuminuric diabetic subjects during 18 months. In the treated group, we observed a reduction of albuminuria from 57.4 mg/24 hours to 35.4 mg/24 hours at the end of the follow up, in contrast with a non significant progression in the group who didn't receive this medication. No significant modification in the clinical and biological parameters was observed during the follow up.
Subject(s)
Albuminuria/drug therapy , Diabetes Complications , Enzyme Inhibitors/pharmacology , Adult , Aged , Albuminuria/etiology , Enzyme Inhibitors/therapeutic use , Female , Humans , Male , Middle Aged , Treatment OutcomeABSTRACT
OBJECTIVE: To assess the effects of insulin therapy on blood pressure levels in type 2 diabetes mellitus (T2 DM). MATERIAL AND METHODS: This is a retrospective analysis of clinical records of 313 T2DM patients (125 men and 188 women), excluding those with proteinuria or hypertensive diseases and those taking drugs that may influence blood pressure levels except antihypertensive therapy. Mean age was 56.3 +/- 11.7 years and mean duration of diabetes was 7.1 +/- 5.5 years. After one week of observation under diet and maximal doses of oral antidiabetic drugs, patients who did not improve their glucose control were changed towards insulin therapy (n=129) and formed the insulin treated group (ITG), those who improved their glucose levels were maintained under oral therapy (n=184) and formed the orally treated group (OTG). Blood pressure levels were compared between the two groups at baseline and after a mean follow-up period of 12.1 +/- 6.1 months. Hypertension was considered if patients were known and treated or if SBP >=140 mmHg and/or DBP >=90 mmHg. RESULTS: At baseline, patients in ITG were moderately older (58.4 +/- 11 vs 54.9 +/- 12.1 years, p<0.05), had a longer duration of diabetes (9.2 +/- 6.2 vs 5.7 +/- 5 years, p<0.01), a lower BMI (24.6 +/- 4.6 vs 28.8 +/- 6.6 kg/m(2), p<0.01) and a higher frequency of retinopathy (44% vs 31.1%, p<0.05). There was no significant difference regarding sex ratio, WHR, family history of hypertension, plasma levels of creatinine and lipid parameters. SBP, DBP and frequency of hypertension were similar in both groups at baseline. After follow up, insulin treated group exhibited higher levels of SBP (150 +/- 25.7 vs 138.6 +/- 27.1 mmHg, p<0.001) and DBP (84.1 +/- 13 vs 75.8 +/- 14.9 mmHg, p<0.001) than orally treated group. Progression rate of hypertension frequency was mildly but not significantly higher in ITG than in OTG (+21% vs +12%, p=0.08) and was associated with weight gain in ITG only. SBP increase was mildly correlated with weight gain (p=0.06). In ITG, higher values of BMI (> 27 kg/m(2)) at baseline were associated with the highest increases of blood pressure levels under insulin therapy. No significant relationship was found with insulin doses. CONCLUSION: Insulin therapy may contribute to the development of hypertension in T2DM obese patients. Additional prospective randomised studies are required for a better appreciation of such influence.
Subject(s)
Diabetes Mellitus, Type 2/drug therapy , Hypertension/chemically induced , Insulin/adverse effects , Adult , Aged , Body Constitution , Body Mass Index , Diabetes Mellitus, Type 2/complications , Diabetic Angiopathies/complications , Diabetic Retinopathy/complications , Female , Humans , Hypoglycemic Agents/therapeutic use , Insulin/administration & dosage , Male , Middle Aged , Retrospective StudiesABSTRACT
OBJECTIVE: To determine the effectiveness and safety of the routine use of antibiotic prophylaxis in women undergoing cesarean section. METHOD: Four hundred and forty-one women undergoing cesarean sections were randomly assigned either to a single dose of 1 g intravenous cefazolin or placebo after clamping of the umbilical cord. The primary outcome was the development of post-operative febrile morbidity and the secondary outcomes were infection-related complications. RESULT: There were 211 emergency and 230 elective cesarean sections. In the emergency cesarean sections, 34 (30.6%) women developed post-operative febrile morbidity in the placebo group compared to 11 (11%) women in the cefazolin group. This was a statistically significant difference (P = 0.001). Similarly, there were statistically significant differences between the two groups in the development of wound infection (P<0.001), use of therapeutic antibiotics (P = 0.001), and post-operative days in hospital (P = 0.003). No statistically significant differences were detected in the development of other infection-related complications. In the elective cesarean sections, no statistically significant differences were found in post-operative febrile morbidity and infection-related complications. There were no serious side effects related to the use of cefazolin. CONCLUSION: The routine use of a single dose of cefazolin is safe and effective in emergency but not elective cesarean section.
Subject(s)
Antibiotic Prophylaxis , Cefazolin/therapeutic use , Cephalosporins/therapeutic use , Cesarean Section/adverse effects , Postoperative Complications/prevention & control , Adult , Cefazolin/administration & dosage , Cefazolin/adverse effects , Cephalosporins/administration & dosage , Cephalosporins/adverse effects , Double-Blind Method , Elective Surgical Procedures , Female , Fever/etiology , Fever/prevention & control , Humans , Infections/drug therapy , Infections/etiology , Injections, Intravenous , Risk Factors , Surgical Wound Infection/prevention & controlABSTRACT
PIP: Demographic, epidemiological, and socioeconomic factors are the primary reasons behind the varied priorities of many developing countries in detecting and preventing osteoporosis. The problem of detecting and preventing the disease does not assume the same urgency, because of the prevailing epidemiological situation in Tunisia and other developing countries. However, there are no precise data on the incidence of osteoporosis and its progression; thus, the means of detection are considered inadequate. Some of the socioeconomic factors in the control of the disease include the following: high rate of illiteracy; excessive focus of the health system on curative care and insufficient attention to early detection and prevention of the disease; and inadequate and inefficient health insurance systems. The increase in life expectancy and the aging of the population make it necessary for health officials to focus their attention on osteoporosis. Attention must be drawn to the importance of nutrition at the very beginning of life, during gestation, wherein deficiencies at this period lead to numerous diseases in adult life. Ensuring that people have strong bones throughout their life spans is a need that can be addressed through the development of an integrated strategy suited to the epidemiological situation of each country.^ieng
Subject(s)
Health Services Needs and Demand , Health Transition , Osteoporosis/prevention & control , Preventive Health Services/organization & administration , Aged , Aged, 80 and over , Female , Humans , Population Surveillance , Socioeconomic Factors , TunisiaSubject(s)
Diabetic Foot/therapy , Sucrose/administration & dosage , Administration, Topical , Adult , Aged , Aged, 80 and over , Anti-Bacterial Agents/administration & dosage , Diabetes Mellitus, Type 1/complications , Diabetes Mellitus, Type 2/complications , Diabetic Foot/drug therapy , Female , Humans , Hydrogen Peroxide/administration & dosage , Male , Middle Aged , Prospective Studies , Solutions , Time Factors , Wound HealingABSTRACT
A prospective study was carried out to compare the fertilizing capability and pregnancy outcome following intracytoplasmic sperm injection (ICSI) using spermatozoa obtained from ejaculates, or surgically from epididymis or seminiferous tubules. A total of 77 ICSI cycles (one per patient) was included. In all, 28 patients had severe oligoasthenoteratozoospermia, 19 patients had obstructive azoospermia and 30 patients had non-obstructive azoospermia. The main outcome measures were fertilization rate per injected metaphase II oocyte and the clinical pregnancy rate per embryo transferred back to the female recipients. In patients with severe oligoasthenoteratozoospermia, the fertilization and pregnancy rates were 79 and 25 %. In patients with obstructive azoospermia, for whom epididymal spermatozoa were used, these were 75 and 28%, and in the non-obstructive group for which testicular spermatozoa were used for injection, they were 69 and 21% respectively. These rates were not significantly different in the three groups (P = 0.85 and P = 0.14 respectively), suggesting that spermatozoa from the ejaculates and epididymal or testicular biopsies are able to fertilize equally by using ICSI. Live birth per embryo transfer was significantly reduced in patients with non-obstructive azoospermia compared to the other two groups. The high abortion rate (50%) in the group in which testicular spermatozoa were used raises doubts about the developmental competence of such embryos.
Subject(s)
Fertilization in Vitro/methods , Spermatozoa , Biopsy , Cell Separation , Epididymis/cytology , Female , Humans , Male , Oligospermia/pathology , Oligospermia/therapy , Pregnancy , Pregnancy Outcome , Prospective Studies , Semen/cytology , Seminiferous Tubules/pathology , Sperm Motility , Spermatozoa/pathology , Spermatozoa/physiology , Suction , Testis/pathologyABSTRACT
Three Tunisian districts were selected to estimate the incidence of insulin-dependent diabetes mellitus (IDDM): Beja, Monastir and Gafsa. A population-based registry for new cases of IDDM was established in 1990 in these three areas according to WHO DIAMOND project methodology. A local extension of the protocol consisted in the inclusion of children up to 19 years of age. Children with a diagnosis of IDDM discharged from general hospitals and private clinics in these areas were recorded in the corresponding registry. A secondary source of case ascertainment was provided by regional school health centers. The findings of the five-year study showed that 156 cases of IDDM were recorded among children aged 0 to 19 years in the three regions. The degree of ascertainment was estimated at 96%. The global age-adjusted incidence rates were 6.76.100,000(-1) year-1 and 6.95.100,000(-1).year-1 in the 0 to 14- and 0 to 19-year age-groups respectively. Age-adjusted incidence rates were lower in Monastir than in Beja and Gafsa, respectively 4.69, 8.13 and 8.33.100,000(-1).year-1 for subjects aged 0 to 19 years. Incidence rates showed no significant difference by gender but were lower in the 0 to 4- and higher in the 10 to 14-year age groups. No time trend was detected. Sixty-two percent of all cases were diagnosed in the cold season. The incidence rate of IDDM in Tunisia is thus close to that observed in most Mediterranean countries.
Subject(s)
Diabetes Mellitus, Type 1/epidemiology , Adolescent , Adult , Age Factors , Censuses , Child , Child, Preschool , Female , Geography , Humans , Incidence , Infant , Male , Patient Selection , Registries , Sex Factors , Tunisia/epidemiologySubject(s)
Anticholesteremic Agents/therapeutic use , Fenofibrate/therapeutic use , Hypercholesterolemia/drug therapy , Lovastatin/analogs & derivatives , Adult , Aged , Cholesterol, HDL/blood , Cholesterol, LDL/blood , Female , Humans , Hypercholesterolemia/blood , Lovastatin/therapeutic use , Male , Middle Aged , Simvastatin , Triglycerides/bloodABSTRACT
The time-course effects of two diets containing raw (RSF) or heated (HSF) soya-bean flour on the digestive enzyme levels in the pancreas and in pancreatic juice were investigated in growing rats fed, alternately, on RSF or HSF diets for two 4-week periods. These values were compared with those obtained in a control group fed on a casein diet. RSF and HSF diets lowered N balance (84.8 (SE 0.9), 82.6 (SE 0.8) and 79.9 (SE 0.8)% with control, HSF and RSF diets respectively, at the third week). However, they increased protease activities compared with the control diet (3-fold for trypsin (EC 3.4.21.4) and chymotrypsin (EC 3.4.21.1) activities in pancreas contents and outputs with the RSF diet; 2-fold for trypsin in pancreas contents and outputs and by 60% for chymotrypsin contents with the HSF diet). The poorer nutritional N utilization might be attributable to soya-bean flour heat-stable (lectins) and heat-labile components (trypsin inhibitors). The decrease in lipid apparent digestibilities in RSF and HSF diets (97.0 (SE 0.8), 91.1 (SE 0.9) and 90.4 (SE 0.7)% with control, RSF and HSF diets at the seventh week) were correlated with a diminution in apparent lipase (EC 3.1.1.3; measured without addition of exogenous colipase), potential lipase (measured with addition of saturated amounts of exogenous colipase) and colipase activities. Compared with control values, gains in potential and apparent lipase outputs were diminished by nearly 40% and gain in colipase outputs by 60% with RSF and HSF diets. These results show clearly that heated or raw soya-bean flours have a significant inhibitory effect on lipase digestive enzyme activities in the pancreas and in its secretion, which might explain impaired lipid digestibility.