Your browser doesn't support javascript.
loading
Show: 20 | 50 | 100
Results 1 - 3 de 3
Filter
Add more filters










Database
Language
Publication year range
1.
Phys Rev Lett ; 132(5): 057301, 2024 Feb 02.
Article in English | MEDLINE | ID: mdl-38364126

ABSTRACT

We compute how small input perturbations affect the output of deep neural networks, exploring an analogy between deep feed-forward networks and dynamical systems, where the growth or decay of local perturbations is characterized by finite-time Lyapunov exponents. We show that the maximal exponent forms geometrical structures in input space, akin to coherent structures in dynamical systems. Ridges of large positive exponents divide input space into different regions that the network associates with different classes. These ridges visualize the geometry that deep networks construct in input space, shedding light on the fundamental mechanisms underlying their learning capabilities.

2.
Neurogastroenterol Motil ; 23(4): 347-55, e159, 2011 Apr.
Article in English | MEDLINE | ID: mdl-21122031

ABSTRACT

BACKGROUND: Few families with autosomal dominant forms of chronic idiopathic pseudo-obstruction (CIP) have been identified and reported. METHODS: We compared two families by clinical, laboratory, histopathologic, and genealogical investigations. Ten patients (pts) (five women) from two families, A and B, both with a family history suggesting autosomal dominant CIP, were investigated. KEY RESULTS: All pts had chronic diarrhea, nine of ten pts had chronic abdominal pain and seven of ten chronic vomiting. Median age for onset of symptoms was 23 (A) and 34 years (B). None had dysphagia, urogenital, neurologic, or ocular symptoms. Small bowel transit and jejunal culture were abnormal in eight of nine. Manometry showed severe jejunal hypomotility in the fasting and fed state and absence of normal phase III in all nine pts and neuropathy-like duodenal alterations in eight of nine. Progress to overt CIP had occurred in six pts. Histopathologic re-evaluation (three pts) showed that criteria of visceral degenerative neuropathy were fulfilled in both families including intranuclear inclusions in all three pts. Genealogic exploration using the unique Swedish Register for Catechetical Meetings disclosed that the two families with all likelihood shared a male ancestor in the 1890 s. CONCLUSIONS & INFERENCES: The compiled results with striking similarities between family A and B together with genealogy findings indicate that this is one, large kindred with a familial autosomal dominant form of intestinal degenerative neuropathy often progressing to CIP but without extra-intestinal manifestations. This is the fourth and, so far, the largest family reported with these characteristics.


Subject(s)
Intestinal Diseases/complications , Intestinal Diseases/genetics , Intestinal Pseudo-Obstruction/etiology , Neurodegenerative Diseases/complications , Neurodegenerative Diseases/genetics , Adolescent , Adult , Aged , Chronic Disease , Disease Progression , Duodenum/pathology , Duodenum/physiopathology , Female , Humans , Intestinal Pseudo-Obstruction/pathology , Intestinal Pseudo-Obstruction/physiopathology , Jejunum/pathology , Jejunum/physiopathology , Male , Manometry , Middle Aged , Pedigree , Sweden , Young Adult
3.
Aliment Pharmacol Ther ; 17(1): 85-92, 2003 Jan.
Article in English | MEDLINE | ID: mdl-12492736

ABSTRACT

BACKGROUND: Systemic glucocorticosteroid therapy is effective in Crohn's disease, but is associated with side-effects. Budesonide has high topical anti-inflammatory activity, but considerably lower systemic activity than other oral glucocorticosteroids. AIM: To evaluate the systemic exposure to budesonide (controlled ileal release capsules) in children and adults with active Crohn's disease, and to assess the suppression of plasma cortisol. METHODS: In an open label study, patients (eight children and six adults) with active Crohn's disease received 9 mg budesonide (Entocort capsules) orally once daily for 7 days. Plasma concentrations were determined on the seventh day of administration, and pharmacokinetic parameters were calculated. For reference, 0.5 mg budesonide was given intravenously separately. Plasma cortisol levels were compared with the pre-treatment baseline values. RESULTS: Systemic exposure to budesonide (AUC0-24 h) after 1 week of oral administration was 41 +/- 21 nmol/L x h (mean +/- s.d.) in children and 35 +/- 20 nmol/L x h in adults. The estimated systemic availability in children was 9 +/- 5% and in adults 11 +/- 7%. The mean plasma cortisol (AUC0-24 h) decreased by 64 +/- 18% in children and by 50 +/- 27% in adults. CONCLUSIONS: The systemic exposure, systemic availability and cortisol suppression after oral administration of 9 mg budesonide were similar in children and adults with active Crohn's disease. Budesonide was well tolerated and no clinically important safety-related findings were identified.


Subject(s)
Anti-Inflammatory Agents/pharmacokinetics , Budesonide/pharmacokinetics , Crohn Disease/drug therapy , Administration, Oral , Adolescent , Adult , Anti-Inflammatory Agents/administration & dosage , Anti-Inflammatory Agents/adverse effects , Budesonide/administration & dosage , Budesonide/adverse effects , Capsules , Child , Crohn Disease/blood , Delayed-Action Preparations , Female , Humans , Hydrocortisone/blood , Infusions, Intravenous , Male
SELECTION OF CITATIONS
SEARCH DETAIL
...